Datasets:

james-burton

/

aug-text-exps-v3

like 0

KNeighborsClassifier

C1

Water Quality Classification

The given case is likely C1 with a confidence level of 87.50% judged based on the values of the input features supplied to the classifier and according to the attributions analysis, F5 and F8 have a high degree of impact. F9, F4, F2, F7, and F3 have a moderate degree of impact while on the contrary F1 and F6 have little impact. Examining further, the values of F5, F8, F9, and F4 all have a positive influence on the classifier supporting the label assignment decision for the given test case. F2 and F3 are also positively supporting features, whereas F7 has a negative influence on the final classification. Finally, F1 and F6 both have very little contributions, though F6 has significantly less than even F1.

{'C1': '87.50%', 'C2': '12.50%'}

{'F5': 'Hardness', 'F8': 'Sulfate', 'F9': 'Solids', 'F4': 'ph', 'F2': 'Organic_carbon', 'F3': 'Conductivity', 'F7': 'Trihalomethanes', 'F1': 'Turbidity', 'F6': 'Chloramines'}

{'F2': 'F5', 'F5': 'F8', 'F3': 'F9', 'F1': 'F4', 'F7': 'F2', 'F6': 'F3', 'F8': 'F7', 'F9': 'F1', 'F4': 'F6'}

{'C2': 'C1', 'C1': 'C2'}

Not Portable

{'C1': 'Not Portable', 'C2': 'Portable'}

RandomForestClassifier

C4

Mobile Price-Range Classification

The label for this example is estimated to be C4 among the four possible classes, with a 73.08 percent chance of being true. C3 is the next most likely label, with a probability of roughly 26.92 percent. The above prediction assessment is mostly dependent on the values of the variables F5, F15, F13, F12, and F14. F5 had the greatest influence, followed by F13, F15, F14, and F12. The positive variables F5, F15, F9, and F1 outnumber the negative variables F13, F14, F12, and F3. Twelve of the twenty variables have values that tilt the prediction towards one of the three other probable classifications. As a result, it is not unexpected that the model is not completely certain of the C4 assigned. Given that the chance of C4's being accurate is 73.08 percent, the model appears to be relatively confident in its final judgement for the data instance under review.

{'C4': '73.08%', 'C3': '26.92%', 'C2': '0.00%', 'C1': '0.00%'}

{'F5': 'ram', 'F13': 'px_width', 'F15': 'battery_power', 'F14': 'px_height', 'F12': 'n_cores', 'F9': 'dual_sim', 'F1': 'touch_screen', 'F3': 'int_memory', 'F10': 'wifi', 'F16': 'fc', 'F18': 'four_g', 'F17': 'm_dep', 'F7': 'pc', 'F8': 'mobile_wt', 'F20': 'talk_time', 'F4': 'three_g', 'F2': 'sc_h', 'F11': 'sc_w', 'F6': 'blue', 'F19': 'clock_speed'}

{'F11': 'F5', 'F10': 'F13', 'F1': 'F15', 'F9': 'F14', 'F7': 'F12', 'F16': 'F9', 'F19': 'F1', 'F4': 'F3', 'F20': 'F10', 'F3': 'F16', 'F17': 'F18', 'F5': 'F17', 'F8': 'F7', 'F6': 'F8', 'F14': 'F20', 'F18': 'F4', 'F12': 'F2', 'F13': 'F11', 'F15': 'F6', 'F2': 'F19'}

{'C2': 'C4', 'C1': 'C3', 'C4': 'C2', 'C3': 'C1'}

r1

{'C4': 'r1', 'C3': 'r2', 'C2': 'r3', 'C1': 'r4'}

BernoulliNB

C1

Personal Loan Modelling

The model has classified the instance as C1 due to the effects of the following features: F2, F6, F3, and F5. Based on the values of these variables, the likelihood of the C1 label is 65.51 percent. F5 and F3 are the top positively contributing variables, whereas F2 and F6 are the most adversely contributing variables. Unlike F5 and F3, which have greater influences on the model's prediction choice in this situation, F4 and F8 have fairly modest positive influences. Finally, F1, F7, and F9 show negative predictive effects, however, as compared to F2, their attributions are modest.

{'C2': '34.49%', 'C1': '65.51%'}

{'F5': 'CD Account', 'F3': 'Income', 'F2': 'CCAvg', 'F6': 'Securities Account', 'F4': 'Education', 'F8': 'Mortgage', 'F1': 'Age', 'F7': 'Family', 'F9': 'Extra_service'}

{'F8': 'F5', 'F2': 'F3', 'F4': 'F2', 'F7': 'F6', 'F5': 'F4', 'F6': 'F8', 'F1': 'F1', 'F3': 'F7', 'F9': 'F9'}

{'C1': 'C2', 'C2': 'C1'}

Accept

{'C2': 'Reject', 'C1': 'Accept'}

DecisionTreeClassifier

C2

Insurance Churn

Considering the predicted likelihoods across the classes, C2 is confidently chosen as the true label since its likelihood is 93.27%, implying that the likelihood of C1 is only about 6.73%. F14 and F9 are the two features with a very strong positive influence, favouring the prediction of class C2. The following features have a moderate effect and are listed in descending order of influence: F3 and F11 have a negative effect, while F7 and F4 have a positive effect on the prediction of C2. Similar to F3 and F11, the features F12 and F1 also negatively affected the prediction decision. Finally, the values of F5, F10, F6, and F16 are the least important to the model decision for this case.

{'C1': '6.73%', 'C2': '93.27%'}

{'F14': 'feature15', 'F9': 'feature14', 'F3': 'feature10', 'F11': 'feature11', 'F7': 'feature5', 'F4': 'feature13', 'F12': 'feature4', 'F1': 'feature3', 'F2': 'feature12', 'F8': 'feature1', 'F13': 'feature7', 'F15': 'feature2', 'F5': 'feature6', 'F10': 'feature0', 'F6': 'feature9', 'F16': 'feature8'}

{'F9': 'F14', 'F8': 'F9', 'F4': 'F3', 'F5': 'F11', 'F15': 'F7', 'F7': 'F4', 'F14': 'F12', 'F13': 'F1', 'F6': 'F2', 'F11': 'F8', 'F1': 'F13', 'F12': 'F15', 'F16': 'F5', 'F10': 'F10', 'F3': 'F6', 'F2': 'F16'}

{'C1': 'C1', 'C2': 'C2'}

Leave

{'C1': 'Stay', 'C2': 'Leave'}

GradientBoostingClassifier

C2

Basketball Players Career Length Prediction

The classification output is C2, however, the classifier is somewhat unsure about this prediction decision because the corresponding predicted probability is only 55.19%. F16 is by far the most influential feature whereas F10, F9, and F14 have been recognised as having the biggest effect on prediction output here after F16. The combination of F16, F10, F9, F14, and F3 features has resulted in the classification choice being altered from C2 to C1. While F18, F17, and F11 all have a minor influence on the classification, F18 is the only one that has a positive impact on the C2 classification. In this case, many features had lower influence on the prediction, with F15, F1, F19, F2, and F8 having a marginal effect.

{'C1': '44.81%', 'C2': '55.19%'}

{'F16': 'GamesPlayed', 'F10': 'OffensiveRebounds', 'F9': 'FieldGoalPercent', 'F14': 'FreeThrowPercent', 'F3': '3PointPercent', 'F18': '3PointAttempt', 'F17': 'FieldGoalsMade', 'F11': 'Blocks', 'F6': 'DefensiveRebounds', 'F7': 'Turnovers', 'F5': 'Rebounds', 'F12': 'MinutesPlayed', 'F4': 'FreeThrowAttempt', 'F13': '3PointMade', 'F15': 'Assists', 'F1': 'PointsPerGame', 'F19': 'FreeThrowMade', 'F2': 'FieldGoalsAttempt', 'F8': 'Steals'}

{'F1': 'F16', 'F13': 'F10', 'F6': 'F9', 'F12': 'F14', 'F9': 'F3', 'F8': 'F18', 'F4': 'F17', 'F18': 'F11', 'F14': 'F6', 'F19': 'F7', 'F15': 'F5', 'F2': 'F12', 'F11': 'F4', 'F7': 'F13', 'F16': 'F15', 'F3': 'F1', 'F10': 'F19', 'F5': 'F2', 'F17': 'F8'}

{'C2': 'C1', 'C1': 'C2'}

Less than 5

{'C1': 'More than 5', 'C2': 'Less than 5'}

LogisticRegression

C1

Customer Churn Modelling

Judging based on the values of the input variables, the classification algorithm labels the case as C1 since its prediction likelihood is equal to 88.69%. The prediction decision is primarily based on the contributions of F10, F6, and F9, however, F2, F4, and F3 are shown to be the least important variables. Regarding the direction of influence of the variables, F10, F9, F5, F2, and F4 are the positive variables that increase the odds of C1 being the correct label. Driving the prediction toward the alternative label, C2, are the variables F6, F1, F7, F8, and F3. Owing to the fact that the most influential variables, F10 and F9, have strong positive attributions, outweighing the contributions of the negative variables, it is not surprising that the algorithm is certain about the decision made.

{'C1': '88.69%', 'C2': '11.31%'}

{'F10': 'IsActiveMember', 'F9': 'NumOfProducts', 'F6': 'Geography', 'F1': 'Gender', 'F7': 'Age', 'F8': 'CreditScore', 'F5': 'EstimatedSalary', 'F2': 'Balance', 'F4': 'Tenure', 'F3': 'HasCrCard'}

{'F9': 'F10', 'F7': 'F9', 'F2': 'F6', 'F3': 'F1', 'F4': 'F7', 'F1': 'F8', 'F10': 'F5', 'F6': 'F2', 'F5': 'F4', 'F8': 'F3'}

{'C2': 'C1', 'C1': 'C2'}

Stay

{'C1': 'Stay', 'C2': 'Leave'}

BernoulliNB

C1

Water Quality Classification

The classification algorithm predicts class C1 with a confidence level of 61.55% and this implies that the probability of the alternative label is only 38.45%. In this case, the top features driving the prediction decision are F8, F9, F2, and F7, followed by F5, F1, F3, F4, and finally F6. Based on the inspections performed to understand the direction of influence of the input features, it can be concluded that F8 has the strongest positive contribution, while F2 has the strongest negative contribution and conversely, all the remaining features have moderate contributions. The other positive features are F9, F5, F1, and F4, whereas the remaining negatives are F7, F3, and F6. All things considered, the influence of the negative features indicates that the likelihood of the C2 label is 38.45% while the positive contributions push the prediction higher towards C1 resulting in the 61.55% prediction confidence.

{'C1': '61.55%', 'C2': '38.45%'}

{'F8': 'Sulfate', 'F9': 'ph', 'F2': 'Trihalomethanes', 'F7': 'Chloramines', 'F5': 'Organic_carbon', 'F1': 'Hardness', 'F3': 'Solids', 'F4': 'Turbidity', 'F6': 'Conductivity'}

{'F5': 'F8', 'F1': 'F9', 'F8': 'F2', 'F4': 'F7', 'F7': 'F5', 'F2': 'F1', 'F3': 'F3', 'F9': 'F4', 'F6': 'F6'}

{'C1': 'C1', 'C2': 'C2'}

Not Portable

{'C1': 'Not Portable', 'C2': 'Portable'}

RandomForestClassifier

C2

Flight Price-Range Classification

The classification model's decision about the true label for the case is based on the information provided to it. Among the three labels, C2, C1, and C3, the model shows without a doubt that neither C1 nor C3 is the true label, given that the probability of C2 being the true label is 100.0%. F11, F6, and F5 are the main contributing factors or variables in the final verdict here since their respective influence outranks the remaining variables. In fact, analysis indicates that F1, F3, and F8 are the least influential variables since they receive little emphasis from the model when making the labelling decision here. In between F11, F6, and F5, and F1, F3, F9, and F8, are the variables such as F4, F2, F12, and F7 with moderate influence on the classification decision here. Among the variables passed to the model, only F4, F7, and F1 are shown to have negative contributions, which suggests that perhaps the true label could be either of the remaining labels. However, given the 100.0% predicted likelihood of C2, it is reasonable to deduce that the positive variables, such as F11, F6, F5, F2, F10, and F12, significantly influence the model's judgement towards C2.

{'C2': '100.00%', 'C1': '0.00%', 'C3': '0.00%'}

{'F11': 'Duration_hours', 'F6': 'Airline', 'F5': 'Total_Stops', 'F4': 'Journey_day', 'F2': 'Source', 'F12': 'Duration_mins', 'F7': 'Arrival_hour', 'F10': 'Destination', 'F9': 'Arrival_minute', 'F1': 'Dep_minute', 'F3': 'Journey_month', 'F8': 'Dep_hour'}

{'F7': 'F11', 'F9': 'F6', 'F12': 'F5', 'F1': 'F4', 'F10': 'F2', 'F8': 'F12', 'F5': 'F7', 'F11': 'F10', 'F6': 'F9', 'F4': 'F1', 'F2': 'F3', 'F3': 'F8'}

{'C2': 'C2', 'C3': 'C1', 'C1': 'C3'}

Low

{'C2': 'Low', 'C1': 'Moderate', 'C3': 'High'}

LogisticRegression

C2

Basketball Players Career Length Prediction

According to the model, C2 is the class with the higher probability, which is equal to 52.57 percent, of being the label for this selected instance or case. Conversely, there is a 47.43 percent chance that C1 is the correct label showing that the model is less certain about the classification verdict in this case. This uncertainty can be linked to the fact that the majority of variables have values that favour assigning C1. The only variables increasing the model's response to prediction C2 are the positive variables namely: F5, F3, F1, F4, F14, F18, and F16. The top negative variables decreasing the likelihood of C2 are F7 and F8 supported by other negative variables, F9, F11, and F13, that further shift the verdict towards C1.

{'C1': '47.43%', 'C2': '52.57%'}

{'F7': '3PointMade', 'F5': '3PointAttempt', 'F8': 'FreeThrowMade', 'F3': 'FreeThrowAttempt', 'F9': 'GamesPlayed', 'F11': 'OffensiveRebounds', 'F13': 'FieldGoalsAttempt', 'F1': 'DefensiveRebounds', 'F2': 'Assists', 'F4': 'MinutesPlayed', 'F14': 'FieldGoalsMade', 'F12': 'Blocks', 'F19': 'Rebounds', 'F15': 'FieldGoalPercent', 'F18': 'Steals', 'F17': 'PointsPerGame', 'F10': 'FreeThrowPercent', 'F16': 'Turnovers', 'F6': '3PointPercent'}

{'F7': 'F7', 'F8': 'F5', 'F10': 'F8', 'F11': 'F3', 'F1': 'F9', 'F13': 'F11', 'F5': 'F13', 'F14': 'F1', 'F16': 'F2', 'F2': 'F4', 'F4': 'F14', 'F18': 'F12', 'F15': 'F19', 'F6': 'F15', 'F17': 'F18', 'F3': 'F17', 'F12': 'F10', 'F19': 'F16', 'F9': 'F6'}

{'C2': 'C1', 'C1': 'C2'}

Less than 5

{'C1': 'More than 5', 'C2': 'Less than 5'}

RandomForestClassifier

C2

Printer Sales

According to the predicted likelihoods across the classes, C1 has a 17.0% chance of being the true label for the given data or case, implying that C2 is the most likely label. F24, F11, and F18 are the most important factors that led to the classification judgments above. The remaining factors have a minor or non-existent impact on the classifier. The classifier most likely ignored the values of F5, F16, F20, F23, F17, and F4 when giving a label to this case since their relative degrees of impact are extremely near to zero. F21, F7, F19, F10, F6, and F12 are considered negative factors among the significant factors because their contributions to the choice tend to reduce the chance that C2 is the correct label. These negatives features lend themselves to the case being classified as C1 but the remaining features contribute positively, raising the likelihood of the C2 classification.

{'C2': '83.00%', 'C1': '17.00%'}

{'F24': 'X8', 'F11': 'X24', 'F18': 'X1', 'F9': 'X2', 'F15': 'X10', 'F13': 'X15', 'F21': 'X25', 'F14': 'X23', 'F26': 'X18', 'F7': 'X4', 'F19': 'X7', 'F25': 'X17', 'F22': 'X3', 'F8': 'X22', 'F10': 'X5', 'F6': 'X9', 'F12': 'X12', 'F1': 'X19', 'F3': 'X11', 'F2': 'X16', 'F5': 'X14', 'F16': 'X21', 'F20': 'X20', 'F23': 'X13', 'F17': 'X6', 'F4': 'X26'}

{'F8': 'F24', 'F24': 'F11', 'F1': 'F18', 'F2': 'F9', 'F10': 'F15', 'F15': 'F13', 'F25': 'F21', 'F23': 'F14', 'F18': 'F26', 'F4': 'F7', 'F7': 'F19', 'F17': 'F25', 'F3': 'F22', 'F22': 'F8', 'F5': 'F10', 'F9': 'F6', 'F12': 'F12', 'F19': 'F1', 'F11': 'F3', 'F16': 'F2', 'F14': 'F5', 'F21': 'F16', 'F20': 'F20', 'F13': 'F23', 'F6': 'F17', 'F26': 'F4'}

{'C2': 'C2', 'C1': 'C1'}

Less

{'C2': 'Less', 'C1': 'More'}

RandomForestClassifier

C2

Credit Risk Classification

According to the ML model, C2 is the most likely class label, and we can conclude that the model is quite confident about the decision given that the probability of having C1 as the correct label is only 7.0%. For the case under study, analysis indicates that F9, F8, F11, and F1 are essentially the negative set of features that push the forecast higher towards C1 instead of C2, while F4, F5, F2, and F7 increase the odds of the prediction being equal to C2. In general, the most relevant feature is F4, while F3 and F10 are the least relevant features, with marginal influence on the above classification verdict. In summary, given the very strong positive influence of F4 together with the moderate influence of the other positives, F5, F7, and F2, it is not strange that the model chose to label the case as C2 instead of C1.

{'C2': '93.00%', 'C1': '7.00%'}

{'F4': 'fea_4', 'F11': 'fea_10', 'F5': 'fea_8', 'F9': 'fea_7', 'F7': 'fea_2', 'F2': 'fea_3', 'F8': 'fea_5', 'F1': 'fea_1', 'F6': 'fea_9', 'F10': 'fea_6', 'F3': 'fea_11'}

{'F4': 'F4', 'F10': 'F11', 'F8': 'F5', 'F7': 'F9', 'F2': 'F7', 'F3': 'F2', 'F5': 'F8', 'F1': 'F1', 'F9': 'F6', 'F6': 'F10', 'F11': 'F3'}

{'C1': 'C2', 'C2': 'C1'}

Low

{'C2': 'Low', 'C1': 'High'}

MLPClassifier

C1

Annual Income Earnings

Because the confidence level associated with the other class, C2, is just 2.29%, the model predicts that the given example is likely C1 and to be specific, the model is quite certain that the right label for the given case is C1. All the features are shown to have some degree of influence on the decision above, with F12 and F13 being the least relevant features, while F10 and F5 are the top features. From the analysis performed to understand how each feature contributes to the above prediction assertion, only the features F4, F9, F8, F3, F11, and F13, have negative influences, shifting the prediction verdict towards C2. The remaining features all contribute positively, strongly shifting the prediction towards the assigned label which could explain the prediction confidence level associated with label C1. The most positive features are F5, F7, and F10 with stronger push in favour of the output label and they are supported by other positive features such as F14, F2, F6, and F1 have a moderate degree of influence.

{'C2': '2.29%', 'C1': '97.71%'}

{'F10': 'Capital Gain', 'F5': 'Marital Status', 'F4': 'Capital Loss', 'F7': 'Relationship', 'F9': 'Hours per week', 'F6': 'Education', 'F1': 'Country', 'F14': 'Age', 'F2': 'Occupation', 'F3': 'Sex', 'F8': 'Education-Num', 'F11': 'Workclass', 'F12': 'fnlwgt', 'F13': 'Race'}

{'F11': 'F10', 'F6': 'F5', 'F12': 'F4', 'F8': 'F7', 'F13': 'F9', 'F4': 'F6', 'F14': 'F1', 'F1': 'F14', 'F7': 'F2', 'F10': 'F3', 'F5': 'F8', 'F2': 'F11', 'F3': 'F12', 'F9': 'F13'}

{'C1': 'C2', 'C2': 'C1'}

Above 50K

{'C2': 'Under 50K', 'C1': 'Above 50K'}

KNNClassifier

C1

Car Acceptability Valuation

The classifier made the prediction here based on the information provided about the case under consideration, and according to the classifier, the prediction probabilities or likelihoods across the labels C1 and C2 are 100.0% and 0.0%, respectively. All the input features are shown to have different degrees of influence on the final decision here by the classifier. The most influential features are F3 and F6, with F1 and F4 ranked as the least contributing factors. The values of F2 and F5 suggest that perhaps the true label could be C2 since they are the negative features. However, considering the confidence in C1, it is valid to conclude that the joint influence or contribution to the classification of the negative features with respect to the given case is outmatched by the joint positive attribution of F3, F6, F1, and F4.

{'C1': '100.00%', 'C2': '0.00%'}

{'F3': 'persons', 'F6': 'safety', 'F5': 'lug_boot', 'F2': 'buying', 'F1': 'doors', 'F4': 'maint'}

{'F4': 'F3', 'F6': 'F6', 'F5': 'F5', 'F1': 'F2', 'F3': 'F1', 'F2': 'F4'}

{'C1': 'C1', 'C2': 'C2'}

Unacceptable

{'C1': 'Unacceptable', 'C2': 'Acceptable'}

LogisticRegression

C1

Real Estate Investment

For the selected case, the model assigns the label C1. The prediction probability distribution across the classes C2 and C1 is 2.40% and 97.60%, respectively. The most important features considered for this prediction are F4, F18, F16, and F17, while on the other hand, the least relevant features with little contributions to the decision based on the analysis are F2, F19, F8, and F5. The top positive features Increasing the likelihood of the prediction being made are F4, F18, and F17. Pushing the prediction towards the alternative class C2, the top negative features are F16, F14, and F20. F3, F11, F15, F10, and F7 are some of the features that have a moderate impact on the classification decision in this case.

{'C2': '2.40%', 'C1': '97.60%'}

{'F4': 'Feature7', 'F18': 'Feature4', 'F16': 'Feature2', 'F17': 'Feature14', 'F14': 'Feature15', 'F20': 'Feature8', 'F3': 'Feature20', 'F11': 'Feature1', 'F15': 'Feature17', 'F7': 'Feature3', 'F10': 'Feature16', 'F12': 'Feature18', 'F13': 'Feature10', 'F1': 'Feature5', 'F9': 'Feature6', 'F6': 'Feature12', 'F2': 'Feature19', 'F19': 'Feature13', 'F8': 'Feature9', 'F5': 'Feature11'}

{'F11': 'F4', 'F9': 'F18', 'F1': 'F16', 'F17': 'F17', 'F4': 'F14', 'F3': 'F20', 'F20': 'F3', 'F7': 'F11', 'F6': 'F15', 'F8': 'F7', 'F18': 'F10', 'F19': 'F12', 'F13': 'F13', 'F2': 'F1', 'F10': 'F9', 'F15': 'F6', 'F5': 'F2', 'F16': 'F19', 'F12': 'F8', 'F14': 'F5'}

{'C2': 'C2', 'C1': 'C1'}

Invest

{'C2': 'Ignore', 'C1': 'Invest'}

MLPClassifier

C2

Vehicle Insurance Claims

The given instance was labelled as C2 by the model based on the values of its features. The model is about 79.64% certain about this prediction decision, hence, there is a slight chance that the label could be C1. Among the different features, the ones with the most impact on the model are F27, F17, F25, F19, and F11. The most negative feature is F27, and it is significantly pushing the narrative toward the prediction of C1. From this, it is foreseeable that there is a chance that the true label could be C1 which is about 20.36%. The influence of F27 and F25 is somewhat counterbalanced by the values of the features F17, F19, and F11. Other attributes that shift the decision in favour of C1 are F30 and F23. F13 shifts the decision further in the direction of C2 and in addition, F16 supports the model's prediction while the values of F9 and F31 of the given test case contradict the model's decision, decreasing the likelihood of C2. Among the features not relevant to this prediction decision for this case are F28, F2, F22, and F10.

{'C2': '79.64%', 'C1': '20.36%'}

{'F27': 'incident_severity', 'F17': 'insured_hobbies', 'F25': 'insured_relationship', 'F19': 'umbrella_limit', 'F11': 'insured_education_level', 'F30': 'authorities_contacted', 'F23': 'incident_type', 'F13': 'policy_csl', 'F16': 'number_of_vehicles_involved', 'F9': 'capital-loss', 'F31': 'property_damage', 'F6': 'insured_occupation', 'F5': 'age', 'F20': 'incident_state', 'F7': 'insured_zip', 'F4': 'collision_type', 'F15': 'property_claim', 'F14': 'injury_claim', 'F12': 'capital-gains', 'F33': 'witnesses', 'F28': 'incident_city', 'F2': 'police_report_available', 'F10': 'months_as_customer', 'F22': 'auto_year', 'F3': 'insured_sex', 'F1': 'policy_state', 'F32': 'vehicle_claim', 'F21': 'total_claim_amount', 'F26': 'bodily_injuries', 'F24': 'incident_hour_of_the_day', 'F18': 'policy_annual_premium', 'F8': 'policy_deductable', 'F29': 'auto_make'}

{'F27': 'F27', 'F23': 'F17', 'F24': 'F25', 'F5': 'F19', 'F21': 'F11', 'F28': 'F30', 'F25': 'F23', 'F19': 'F13', 'F10': 'F16', 'F8': 'F9', 'F31': 'F31', 'F22': 'F6', 'F2': 'F5', 'F29': 'F20', 'F6': 'F7', 'F26': 'F4', 'F15': 'F15', 'F14': 'F14', 'F7': 'F12', 'F12': 'F33', 'F30': 'F28', 'F32': 'F2', 'F1': 'F10', 'F17': 'F22', 'F20': 'F3', 'F18': 'F1', 'F16': 'F32', 'F13': 'F21', 'F11': 'F26', 'F9': 'F24', 'F4': 'F18', 'F3': 'F8', 'F33': 'F29'}

{'C1': 'C2', 'C2': 'C1'}

Not Fraud

{'C2': 'Not Fraud', 'C1': 'Fraud'}

RandomForestClassifier

C2

Ethereum Fraud Detection

According to the classification algorithm, the best label for the given case is C2, because there is little to no chance that C1 is the correct label. Not all of the features are found to contribute to the label given here. The following significant features are ordered in order of their effect on the algorithm's output: F33, F12, F8, F37, F18, F25, F21, F13, F20, F24, F16, F14, F2, F32, F30, F38, F35, F6, F26, F4. F3, F5, and F9, on the other hand, are unimportant features since they have almost no influence. Among the most influential features F33, F12, F8, F37, and F18, F8 is considered the most negative, dragging the verdict in a different direction, while the others have positive contributions, increasing the possibility that C2 is correct in this case. F21 is recognised as a positive feature with modest effect, whereas F25 and F13 are identified as negative features. Given that the majority of the top five attributes have positive contributions, boosting the likelihood that C2 is the correct label, it is not unexpected that the algorithm is quite confident in the assigned label's accuracy.

{'C1': '0.00%', 'C2': '100.00%'}

{'F33': ' ERC20 total Ether sent contract', 'F12': ' ERC20 min val rec', 'F8': 'total transactions (including tnx to create contract', 'F37': ' ERC20 max val rec', 'F18': ' Total ERC20 tnxs', 'F25': ' ERC20 uniq rec addr', 'F21': 'min val sent', 'F13': 'Time Diff between first and last (Mins)', 'F20': 'Sent tnx', 'F24': 'Avg min between received tnx', 'F16': 'min value received', 'F14': ' ERC20 total ether sent', 'F2': 'avg val sent', 'F32': 'max val sent', 'F30': 'Avg min between sent tnx', 'F38': 'Received Tnx', 'F35': ' ERC20 uniq sent token name', 'F6': 'Unique Sent To Addresses', 'F26': ' ERC20 uniq rec token name', 'F4': ' ERC20 uniq rec contract addr', 'F3': 'total Ether sent', 'F5': 'Number of Created Contracts', 'F9': ' ERC20 avg val sent', 'F31': ' ERC20 max val sent', 'F1': ' ERC20 min val sent', 'F27': ' ERC20 avg val rec', 'F36': 'Unique Received From Addresses', 'F19': 'max value received ', 'F28': ' ERC20 uniq sent addr.1', 'F29': 'total ether sent contracts', 'F15': 'avg val received', 'F10': ' ERC20 uniq sent addr', 'F17': 'min value sent to contract', 'F11': 'max val sent to contract', 'F7': ' ERC20 total Ether received', 'F34': 'avg value sent to contract', 'F22': 'total ether balance', 'F23': 'total ether received'}

{'F26': 'F33', 'F31': 'F12', 'F18': 'F8', 'F32': 'F37', 'F23': 'F18', 'F28': 'F25', 'F12': 'F21', 'F3': 'F13', 'F4': 'F20', 'F2': 'F24', 'F9': 'F16', 'F25': 'F14', 'F14': 'F2', 'F13': 'F32', 'F1': 'F30', 'F5': 'F38', 'F37': 'F35', 'F8': 'F6', 'F38': 'F26', 'F30': 'F4', 'F19': 'F3', 'F6': 'F5', 'F36': 'F9', 'F35': 'F31', 'F34': 'F1', 'F33': 'F27', 'F7': 'F36', 'F10': 'F19', 'F29': 'F28', 'F21': 'F29', 'F11': 'F15', 'F27': 'F10', 'F15': 'F17', 'F16': 'F11', 'F24': 'F7', 'F17': 'F34', 'F22': 'F22', 'F20': 'F23'}

{'C1': 'C1', 'C2': 'C2'}

Fraud

{'C1': 'Not Fraud', 'C2': 'Fraud'}

BernoulliNB

C1

Hotel Satisfaction

The classifier labbelled the given case as C1 with a confidence level of 98.89%, implying that the chance of C2 being the correct label is only about 1.11%. The classification output decision is solely based on the information supplied to the classifier about the case under review. We can rank the contributions of the features as follows: F6, F9, F14, F12, F10, F11, F8, F7, F4, F15, F3, F5, F1, F2, and F13. Among the top features, F6 is the only negative feature, increasing the probability of predicting the alternative label, C2. Other top features that are shifting the prediction towards C1 are F9, F14, and F12. Similar to F6, the features F8, F2, and F15 have negative contributions, supporting the generation of C2. By comparing the strong joint positive attribution to the joint negative attribution, it is evident why the classifier is very certain that C1 is the right label for this instance.

{'C1': '98.89%', 'C2': '1.11%'}

{'F6': 'Type of Travel', 'F9': 'Type Of Booking', 'F14': 'Common Room entertainment', 'F12': 'Stay comfort', 'F10': 'Cleanliness', 'F11': 'Hotel wifi service', 'F8': 'Other service', 'F7': 'Ease of Online booking', 'F4': 'Age', 'F15': 'Checkin\\/Checkout service', 'F3': 'Food and drink', 'F5': 'Departure\\/Arrival convenience', 'F1': 'purpose_of_travel', 'F2': 'Hotel location', 'F13': 'Gender'}

{'F3': 'F6', 'F4': 'F9', 'F12': 'F14', 'F11': 'F12', 'F15': 'F10', 'F6': 'F11', 'F14': 'F8', 'F8': 'F7', 'F5': 'F4', 'F13': 'F15', 'F10': 'F3', 'F7': 'F5', 'F2': 'F1', 'F9': 'F2', 'F1': 'F13'}

{'C2': 'C1', 'C1': 'C2'}

dissatisfied

{'C1': 'dissatisfied', 'C2': 'satisfied'}

RandomForestClassifier

C1

Used Cars Price-Range Prediction

The prediction probability associated with class C2 is 10.50%, while that of class C1 is 89.50%, therefore, it can be concluded that C1 is the most probable label for the given case according to the model. All the input features are shown to contribute to the above decision, and the ones with the strongest influence on the classification decision are F2, F7, and F6, but F8, F10, F9, and F5 are shown to be the least relevant features . Finally, the degree of influence of F4, F1, and F3 can be described as moderate. The model's high confidence can be attributed to the strong positive contributions of F7 and F2 which are supported by the contributions of the remaining positive features F4, F8, and F10. Conversely, shifting the prediction in favour of C2, the negative features F6, F1, F9, F3, and F5.

{'C2': '10.50%', 'C1': '89.50%'}

{'F7': 'Power', 'F2': 'car_age', 'F6': 'Transmission', 'F4': 'Fuel_Type', 'F1': 'Name', 'F3': 'Mileage', 'F8': 'Engine', 'F9': 'Owner_Type', 'F10': 'Kilometers_Driven', 'F5': 'Seats'}

{'F4': 'F7', 'F5': 'F2', 'F8': 'F6', 'F7': 'F4', 'F6': 'F1', 'F2': 'F3', 'F3': 'F8', 'F9': 'F9', 'F1': 'F10', 'F10': 'F5'}

{'C1': 'C2', 'C2': 'C1'}

High

{'C2': 'Low', 'C1': 'High'}

SVC

C1

Food Ordering Customer Churn Prediction

The model labels the case as C1 with fairly high confidence equal to 89.73%, whereas the likelihood of C2 is only 10.27%. Analysis shows that only 20 of the 46 input variables contribute to the prediction assertion above. The prediction judgement C1 is mainly based on the variables F16, F29, F23, and F1. F22, F32, F25, F21, F40, and F36 also contribute to the decision, however, their degree of influence is only moderate. According to the direction of influence analysis, F16, F1, F40, and F21 positively support the decision of the model to assign the label C1. However, F29, F32, F36, F23, F22, and F25 reduce the likelihood or chance that C1 is the true label for this particular test instance. The main variables with less influence on the above classification decision are F10, F20, F44, and F45.

{'C1': '89.73%', 'C2': '10.27%'}

{'F16': 'Ease and convenient', 'F29': 'Unaffordable', 'F1': 'Good Food quality', 'F23': 'Wrong order delivered', 'F22': 'Delay of delivery person picking up food', 'F32': 'Politeness', 'F25': 'Self Cooking', 'F36': 'Late Delivery', 'F21': 'Health Concern', 'F40': 'More Offers and Discount', 'F43': 'Easy Payment option', 'F18': 'Time saving', 'F4': 'Perference(P2)', 'F6': 'Gender', 'F8': 'Good Road Condition', 'F17': 'Google Maps Accuracy', 'F13': 'Good Taste ', 'F9': 'Good Tracking system', 'F5': 'Bad past experience', 'F14': 'Marital Status', 'F10': 'Influence of rating', 'F44': 'Delivery person ability', 'F20': 'Low quantity low time', 'F45': 'Age', 'F7': 'Less Delivery time', 'F38': 'High Quality of package', 'F28': 'Maximum wait time', 'F34': 'Number of calls', 'F39': 'Freshness ', 'F41': 'Temperature', 'F37': 'Residence in busy location', 'F3': 'Long delivery time', 'F42': 'Order Time', 'F27': 'Influence of time', 'F12': 'Order placed by mistake', 'F31': 'Missing item', 'F19': 'Delay of delivery person getting assigned', 'F46': 'Family size', 'F26': 'Unavailability', 'F35': 'Poor Hygiene', 'F33': 'More restaurant choices', 'F24': 'Perference(P1)', 'F30': 'Educational Qualifications', 'F2': 'Monthly Income', 'F15': 'Occupation', 'F11': 'Good Quantity'}

{'F10': 'F16', 'F23': 'F29', 'F15': 'F1', 'F27': 'F23', 'F26': 'F22', 'F42': 'F32', 'F17': 'F25', 'F19': 'F36', 'F18': 'F21', 'F14': 'F40', 'F13': 'F43', 'F11': 'F18', 'F9': 'F4', 'F2': 'F6', 'F35': 'F8', 'F34': 'F17', 'F45': 'F13', 'F16': 'F9', 'F21': 'F5', 'F3': 'F14', 'F38': 'F10', 'F37': 'F44', 'F36': 'F20', 'F1': 'F45', 'F39': 'F7', 'F40': 'F38', 'F32': 'F28', 'F41': 'F34', 'F43': 'F39', 'F44': 'F41', 'F33': 'F37', 'F24': 'F3', 'F31': 'F42', 'F30': 'F27', 'F29': 'F12', 'F28': 'F31', 'F25': 'F19', 'F7': 'F46', 'F22': 'F26', 'F20': 'F35', 'F12': 'F33', 'F8': 'F24', 'F6': 'F30', 'F5': 'F2', 'F4': 'F15', 'F46': 'F11'}

{'C2': 'C1', 'C1': 'C2'}

Return

{'C1': 'Return', 'C2': 'Go Away'}

MLPClassifier

C2

Annual Income Earnings

The label predicted for this case is C2 with very high confidence of approximately 97.71% which insinuates that there is a marginal possibility that C1 could be the label. The above classification decision is largely due to the values of F1, F12, F10, and F7. On the other hand, F8 and F14 are less relevant when the model is deciding the correct label for the case here. Digging deeper revealed that each feature either positively or negatively contribute to the prediction made here. Six features contradicted the classification decision, while the remaining ones positively supported the C2 prediction. The negative features driving the prediction towards C1 are F10, F4, F6, F13, F5, and F14 and countering their influence are the top positive features are F1, F12, F2, and F7.

{'C1': '2.29%', 'C2': '97.71%'}

{'F1': 'Capital Gain', 'F12': 'Marital Status', 'F10': 'Capital Loss', 'F7': 'Relationship', 'F4': 'Hours per week', 'F2': 'Education', 'F3': 'Country', 'F9': 'Age', 'F11': 'Occupation', 'F6': 'Sex', 'F13': 'Education-Num', 'F5': 'Workclass', 'F8': 'fnlwgt', 'F14': 'Race'}

{'F11': 'F1', 'F6': 'F12', 'F12': 'F10', 'F8': 'F7', 'F13': 'F4', 'F4': 'F2', 'F14': 'F3', 'F1': 'F9', 'F7': 'F11', 'F10': 'F6', 'F5': 'F13', 'F2': 'F5', 'F3': 'F8', 'F9': 'F14'}

{'C2': 'C1', 'C1': 'C2'}

Above 50K

{'C1': 'Under 50K', 'C2': 'Above 50K'}

SVM_linear

C1

Wine Quality Prediction

The likelihood of C1 being the correct label for the selected case or instance is 67.54% according to the classifier. This means, there is a 32.46% chance that C2 could be the label and the classification assertion above is influenced mainly by the variables F9, F1, F8, and F10. On the contrary, F2, F6, and F11 are deemed less important when deciding the correct label for this given case. Decreasing the likelihood of the predicted label , C1, are the variables F10, F4, F6, and F11, therefore, these negative variables support the alternative class C2. However, the collective or joint attribution of the top positive variables, F1, F9, and F8 is strong enough to tilt the classification in favour of C1.

{'C2': '32.46%', 'C1': '67.54%'}

{'F9': 'residual sugar', 'F1': 'volatile acidity', 'F8': 'alcohol', 'F10': 'fixed acidity', 'F7': 'chlorides', 'F4': 'sulphates', 'F3': 'citric acid', 'F5': 'free sulfur dioxide', 'F2': 'density', 'F6': 'total sulfur dioxide', 'F11': 'pH'}

{'F4': 'F9', 'F2': 'F1', 'F11': 'F8', 'F1': 'F10', 'F5': 'F7', 'F10': 'F4', 'F3': 'F3', 'F6': 'F5', 'F8': 'F2', 'F7': 'F6', 'F9': 'F11'}

{'C1': 'C2', 'C2': 'C1'}

high quality

{'C2': 'low_quality', 'C1': 'high quality'}

KNeighborsClassifier

C1

Credit Risk Classification

The confidence level score with respect to each class label suggests that this case should be labelled as C1. Specifically, there is about an 80.0% chance that C1 is the correct label. However, this implies that there is also about a 20.0% chance that it should be C2. The above prediction decision is based predominantly on the influence of the following features: F2, F9, F7, F4, F11, F6, and F1. According to the analysis, the features F2, F9, and F7 have a very strong positive influence, swinging the prediction decision towards C1. In contrast, the value of F4 also suggests the decision should be the alternative class, C2. Similar to F4, the values of F10, F11, and F6 indicate the label could be C2. However, the influence of these features is very small compared to F2, F9, F7, and F4. Finally, the attributes with a moderately low influence on the final prediction decision for this case include F1, F5, F8, and F3. The values of F1 and F3 have a negative attribution, while F5 and F8 have positive attributions.

{'C1': '80.00%', 'C2': '20.00%'}

{'F2': 'fea_4', 'F9': 'fea_8', 'F7': 'fea_2', 'F4': 'fea_9', 'F10': 'fea_6', 'F11': 'fea_10', 'F6': 'fea_1', 'F1': 'fea_11', 'F5': 'fea_7', 'F3': 'fea_3', 'F8': 'fea_5'}

{'F4': 'F2', 'F8': 'F9', 'F2': 'F7', 'F9': 'F4', 'F6': 'F10', 'F10': 'F11', 'F1': 'F6', 'F11': 'F1', 'F7': 'F5', 'F3': 'F3', 'F5': 'F8'}

{'C2': 'C1', 'C1': 'C2'}

Low

{'C1': 'Low', 'C2': 'High'}

BernoulliNB

C1

Job Change of Data Scientists

The classification algorithm is pretty confident that the correct label for the data under consideration is C1, wowever, it is noteworthy to consider that C2 has about a 15.13% chance of being the correct label. The predicted probability of each label is assigned based on the influence of features such as F2, F10, F6, and F11. However, the analysis shows that the values of F8, F1, F3, and F9 are less relevant when classifying the data. Only the features F6, F5, F8, F1, F3, and F9 have negative attributions, decreasing the predicted probability of the assigned label and one can say these features are shifting the prediction decision towards the label C2.

{'C2': '15.13%', 'C1': '84.87%'}

{'F2': 'city', 'F10': 'enrolled_university', 'F6': 'relevent_experience', 'F11': 'city_development_index', 'F5': 'experience', 'F4': 'education_level', 'F7': 'major_discipline', 'F12': 'last_new_job', 'F8': 'gender', 'F1': 'company_size', 'F3': 'company_type', 'F9': 'training_hours'}

{'F3': 'F2', 'F6': 'F10', 'F5': 'F6', 'F1': 'F11', 'F9': 'F5', 'F7': 'F4', 'F8': 'F7', 'F12': 'F12', 'F4': 'F8', 'F10': 'F1', 'F11': 'F3', 'F2': 'F9'}

{'C2': 'C2', 'C1': 'C1'}

Leave

{'C2': 'Stay', 'C1': 'Leave'}

GradientBoostingClassifier

C2

Printer Sales

The case, despite having features with considerable negative impact, also has numerous and measurable positive features, so the assignment of the label C2 by the model is very likely since the predicted probability is 91.95% which is very higher than that of C1. The F23, F19, and F11 were the most important features driving the model to arrive at the labelling assignment of C2. F14 and F26 have nearly identical positive attributions, while F13 and F25 has negative impacts, swinging the prediction towards a different label. However, the joint positive impact of F14, F23, F11, and F26 stands out over the impact of F19, F22, F25, and F13, favouring the prediction of the C2 model. All things considered, there are more features with a positive impact than those with negative impact; the mean attribution of the positive attributes is much larger which somewhat explains why the confidence level is very high. Above all, it is important to note that the prediction is made with less emphasis on the values of F8, F4, F17, and F7 hence they are practically irrelevant when it comes to labelling this case.

{'C1': '8.05%', 'C2': '91.95%'}

{'F23': 'X24', 'F19': 'X8', 'F11': 'X1', 'F22': 'X21', 'F25': 'X4', 'F26': 'X6', 'F14': 'X3', 'F13': 'X22', 'F2': 'X7', 'F6': 'X15', 'F15': 'X20', 'F3': 'X11', 'F24': 'X10', 'F12': 'X19', 'F1': 'X5', 'F18': 'X16', 'F10': 'X23', 'F16': 'X9', 'F9': 'X17', 'F21': 'X18', 'F4': 'X25', 'F20': 'X14', 'F7': 'X2', 'F17': 'X13', 'F5': 'X12', 'F8': 'X26'}

{'F24': 'F23', 'F8': 'F19', 'F1': 'F11', 'F21': 'F22', 'F4': 'F25', 'F6': 'F26', 'F3': 'F14', 'F22': 'F13', 'F7': 'F2', 'F15': 'F6', 'F20': 'F15', 'F11': 'F3', 'F10': 'F24', 'F19': 'F12', 'F5': 'F1', 'F16': 'F18', 'F23': 'F10', 'F9': 'F16', 'F17': 'F9', 'F18': 'F21', 'F25': 'F4', 'F14': 'F20', 'F2': 'F7', 'F13': 'F17', 'F12': 'F5', 'F26': 'F8'}

{'C1': 'C1', 'C2': 'C2'}

More

{'C1': 'Less', 'C2': 'More'}

LogisticRegression

C2

Cab Surge Pricing System

The predicted label is C2 given the predictability of C1 is 28.96% and that of C3 is 23.41%. Considering the probabilities of the classes, the model can be described as being moderately confident. The prediction of C2 can be attributed to the varying degree of contributions of the input features. Attribution analysis indicates that F9, F11, and F10 are considered the most influential. Those with moderate influence are F3, F6, F5, F4, F12, and F2, whereas on the contrary, the least influential ones are F8, F7, and F1. The analysis also revealed that not all the features contribute positively to the prediction decision and amongst the input features, the ones with negative attributions decreasing the likelihood of the C2 prediction are F11, F10, F5, F4, and F12 whereas conversely, the top positive features are F9, F3, and F6.

{'C1': '28.96%', 'C3': '23.41%', 'C2': '47.63%'}

{'F9': 'Type_of_Cab', 'F11': 'Confidence_Life_Style_Index', 'F10': 'Destination_Type', 'F3': 'Trip_Distance', 'F6': 'Cancellation_Last_1Month', 'F5': 'Life_Style_Index', 'F4': 'Customer_Rating', 'F12': 'Var3', 'F2': 'Var1', 'F8': 'Customer_Since_Months', 'F7': 'Var2', 'F1': 'Gender'}

{'F2': 'F9', 'F5': 'F11', 'F6': 'F10', 'F1': 'F3', 'F8': 'F6', 'F4': 'F5', 'F7': 'F4', 'F11': 'F12', 'F9': 'F2', 'F3': 'F8', 'F10': 'F7', 'F12': 'F1'}

{'C3': 'C1', 'C1': 'C3', 'C2': 'C2'}

C3

{'C1': 'Low', 'C3': 'Medium', 'C2': 'High'}

RandomForestClassifier

C1

Company Bankruptcy Prediction

The model outputs a predicted probability of 2.55% for the C2 label and 97.45% for the C1 label. Judging from above, the most probable class is C1. Hence, C1 is the assigned label by the model, with a very high confidence level. The top features contributing to the prediction assessment above are F67, F6, F81, F52, and F51. However, the values of about twenty features are deemed relevant while the remaining are regarded as irrelevant when classifying the given case. These irrelevant features include F80, F48, F91, and F53. Among the relevant features, F81, F43, F90, F70, F42, and F12 are shown to be the only positive features that increase the model's response in favour of the assigned label C1. In contrast, the majority of the relevant features, mainly F67, F6, F52, and F51, have negative contributions, decreasing the odds of the label C1, hence supporting the assignment of C2 to the given case.

{'C2': '2.55%', 'C1': '97.45%'}

{'F67': " Net Income to Stockholder's Equity", 'F6': ' Total income\\/Total expense', 'F81': ' Borrowing dependency', 'F52': ' Continuous interest rate (after tax)', 'F51': ' Net Value Per Share (B)', 'F11': ' Cash\\/Current Liability', 'F54': ' Net worth\\/Assets', 'F43': ' Fixed Assets Turnover Frequency', 'F90': ' Interest-bearing debt interest rate', 'F70': ' No-credit Interval', 'F69': ' Net Value Per Share (A)', 'F44': ' Long-term fund suitability ratio (A)', 'F42': ' Equity to Long-term Liability', 'F21': ' Realized Sales Gross Margin', 'F12': ' Current Asset Turnover Rate', 'F32': ' Working Capital to Total Assets', 'F27': ' Interest Coverage Ratio (Interest expense to EBIT)', 'F60': ' Working capitcal Turnover Rate', 'F83': ' Inventory Turnover Rate (times)', 'F13': ' After-tax net Interest Rate', 'F80': ' Working Capital\\/Equity', 'F48': ' Liability to Equity', 'F91': ' Operating Gross Margin', 'F53': ' Cash Flow Per Share', 'F38': ' Contingent liabilities\\/Net worth', 'F78': ' Operating Profit Per Share (Yuan ¥)', 'F41': ' Operating Profit Rate', 'F25': ' Net Worth Turnover Rate (times)', 'F63': ' Continuous Net Profit Growth Rate', 'F46': ' Long-term Liability to Current Assets', 'F89': ' Fixed Assets to Assets', 'F4': ' Inventory and accounts receivable\\/Net value', 'F17': ' Regular Net Profit Growth Rate', 'F8': ' Current Liability to Equity', 'F66': ' Equity to Liability', 'F73': ' Current Liability to Liability', 'F10': ' Operating profit\\/Paid-in capital', 'F57': ' Net Value Per Share (C)', 'F76': ' Operating Funds to Liability', 'F5': ' Current Liability to Current Assets', 'F86': ' Current Ratio', 'F93': ' Quick Assets\\/Current Liability', 'F75': ' Tax rate (A)', 'F26': ' After-tax Net Profit Growth Rate', 'F7': ' Per Share Net profit before tax (Yuan ¥)', 'F72': ' Total Asset Turnover', 'F40': ' CFO to Assets', 'F23': ' Cash Reinvestment %', 'F47': ' Net profit before tax\\/Paid-in capital', 'F88': ' Cash Flow to Equity', 'F3': ' Debt ratio %', 'F2': ' Current Liabilities\\/Liability', 'F61': ' Interest Expense Ratio', 'F77': ' Cash Flow to Sales', 'F64': ' Total Asset Growth Rate', 'F59': ' Inventory\\/Current Liability', 'F34': ' Allocation rate per person', 'F84': ' Operating Expense Rate', 'F16': ' Operating profit per person', 'F58': ' Net Income to Total Assets', 'F37': ' Net Value Growth Rate', 'F92': ' ROA(B) before interest and depreciation after tax', 'F1': ' Cash Flow to Liability', 'F14': ' Inventory\\/Working Capital', 'F30': ' Retained Earnings to Total Assets', 'F15': ' Total assets to GNP price', 'F24': ' Persistent EPS in the Last Four Seasons', 'F87': ' Total debt\\/Total net worth', 'F62': ' Quick Ratio', 'F71': ' Revenue per person', 'F39': ' Non-industry income and expenditure\\/revenue', 'F28': ' Cash\\/Total Assets', 'F65': ' ROA(A) before interest and % after tax', 'F50': ' ROA(C) before interest and depreciation before interest', 'F82': ' Research and development expense rate', 'F18': ' Cash Flow to Total Assets', 'F45': ' Pre-tax net Interest Rate', 'F36': ' Accounts Receivable Turnover', 'F74': ' Current Liability to Assets', 'F20': ' Quick Assets\\/Total Assets', 'F85': ' Total expense\\/Assets', 'F19': ' Operating Profit Growth Rate', 'F79': ' Average Collection Days', 'F9': ' Current Assets\\/Total Assets', 'F55': ' Current Liabilities\\/Equity', 'F29': ' Realized Sales Gross Profit Growth Rate', 'F35': ' Cash flow rate', 'F56': ' Total Asset Return Growth Rate Ratio', 'F33': ' Degree of Financial Leverage (DFL)', 'F68': ' Cash Turnover Rate', 'F22': ' Quick Asset Turnover Rate', 'F49': ' Revenue Per Share (Yuan ¥)', 'F31': ' Gross Profit to Sales'}

{'F59': 'F67', 'F57': 'F6', 'F3': 'F81', 'F12': 'F52', 'F27': 'F51', 'F32': 'F11', 'F84': 'F54', 'F22': 'F43', 'F1': 'F90', 'F56': 'F70', 'F42': 'F69', 'F52': 'F44', 'F23': 'F42', 'F83': 'F21', 'F61': 'F12', 'F67': 'F32', 'F60': 'F27', 'F73': 'F60', 'F18': 'F83', 'F79': 'F13', 'F68': 'F80', 'F66': 'F48', 'F62': 'F91', 'F65': 'F53', 'F64': 'F38', 'F63': 'F78', 'F58': 'F41', 'F55': 'F25', 'F54': 'F63', 'F69': 'F46', 'F74': 'F89', 'F70': 'F4', 'F85': 'F17', 'F92': 'F8', 'F91': 'F66', 'F90': 'F73', 'F89': 'F10', 'F88': 'F57', 'F87': 'F76', 'F86': 'F5', 'F82': 'F86', 'F71': 'F93', 'F81': 'F75', 'F80': 'F26', 'F78': 'F7', 'F77': 'F72', 'F76': 'F40', 'F75': 'F23', 'F72': 'F47', 'F53': 'F88', 'F47': 'F3', 'F51': 'F2', 'F14': 'F61', 'F25': 'F77', 'F24': 'F64', 'F21': 'F59', 'F20': 'F34', 'F19': 'F84', 'F17': 'F16', 'F16': 'F58', 'F15': 'F37', 'F13': 'F92', 'F50': 'F1', 'F11': 'F14', 'F10': 'F30', 'F9': 'F15', 'F8': 'F24', 'F7': 'F87', 'F6': 'F62', 'F5': 'F71', 'F4': 'F39', 'F26': 'F28', 'F28': 'F65', 'F29': 'F50', 'F30': 'F82', 'F49': 'F18', 'F48': 'F45', 'F2': 'F36', 'F46': 'F74', 'F45': 'F20', 'F44': 'F85', 'F43': 'F19', 'F41': 'F79', 'F40': 'F9', 'F39': 'F55', 'F38': 'F29', 'F37': 'F35', 'F36': 'F56', 'F35': 'F33', 'F34': 'F68', 'F33': 'F22', 'F31': 'F49', 'F93': 'F31'}

{'C1': 'C2', 'C2': 'C1'}

Yes

{'C2': 'No', 'C1': 'Yes'}

SVC

C2

Broadband Sevice Signup

The algorithm identifies the provided data or case as C2 with a greater level of certainty since the prediction probability of class C1 is just 0.07 percent as a result, C1 is less likely than C2. The influence of input features such as F13, F15, F12, F23, and F6 is mostly responsible for the classification verdict above with only F6 having a negative influence among them, slightly pulling the decision in favour of C1. F13, F15, F12, and F23, on the other hand, make considerable positive contributions in favour of assigning C2 to the data. F1, F8, F28, F34, F14, F17, F5, and F16 are some more features that have a modest effect on the algorithm's decision. But, not all features are demonstrated to influence the classification decision either negatively or positively to the aforementioned classification outcome and in reality, a number of these are demonstrated to be irrelevant for determining the suitable label for this case and these include F25, F18, F38, and F29. All in all, the most important features for this classification instance are F13 and F15, whereas F33 and F36 are the least important.

{'C2': '99.93%', 'C1': '0.07%'}

{'F13': 'X38', 'F15': 'X32', 'F12': 'X31', 'F23': 'X25', 'F6': 'X8', 'F1': 'X35', 'F8': 'X1', 'F5': 'X3', 'F34': 'X28', 'F16': 'X19', 'F14': 'X9', 'F28': 'X11', 'F17': 'X10', 'F2': 'X21', 'F4': 'X17', 'F21': 'X4', 'F9': 'X36', 'F37': 'X2', 'F39': 'X6', 'F19': 'X34', 'F18': 'X37', 'F25': 'X40', 'F38': 'X42', 'F29': 'X41', 'F27': 'X5', 'F42': 'X33', 'F20': 'X39', 'F24': 'X24', 'F22': 'X30', 'F10': 'X27', 'F35': 'X26', 'F31': 'X23', 'F3': 'X22', 'F26': 'X20', 'F32': 'X18', 'F7': 'X16', 'F40': 'X15', 'F41': 'X14', 'F36': 'X13', 'F33': 'X12', 'F30': 'X7', 'F11': 'X29'}

{'F35': 'F13', 'F29': 'F15', 'F28': 'F12', 'F23': 'F23', 'F6': 'F6', 'F32': 'F1', 'F40': 'F8', 'F2': 'F5', 'F26': 'F34', 'F17': 'F16', 'F7': 'F14', 'F9': 'F28', 'F8': 'F17', 'F19': 'F2', 'F15': 'F4', 'F3': 'F21', 'F33': 'F9', 'F1': 'F37', 'F4': 'F39', 'F31': 'F19', 'F34': 'F18', 'F37': 'F25', 'F38': 'F38', 'F39': 'F29', 'F41': 'F27', 'F30': 'F42', 'F36': 'F20', 'F22': 'F24', 'F27': 'F22', 'F25': 'F10', 'F24': 'F35', 'F21': 'F31', 'F20': 'F3', 'F18': 'F26', 'F16': 'F32', 'F14': 'F7', 'F13': 'F40', 'F12': 'F41', 'F11': 'F36', 'F10': 'F33', 'F5': 'F30', 'F42': 'F11'}

{'C1': 'C2', 'C2': 'C1'}

No

{'C2': 'No', 'C1': 'Yes'}

SVM

C2

Customer Churn Modelling

For the given dataset instance, the label assigned by the classifier is C2 since it has a predicted probability of about 89.16%. On the other hand, there is a 9.0% chance that C3 could be the appropriate label, whereas C1 only has a 1.84% chance of being the true label. The classifier arrived at this classification verdict chiefly due to the influence and contributions of variables such as F9, F8, F7, and F10. However, there is less emphasis on the values of F1, F6, and F2, since their impact on the classifier with respect to the given case is smaller compared to the other variables, hence they are the least ranked features. From the attribution analysis, there are four variables with negative contributions, pushing the verdict in the direction of C3. These negative variables are F9, F10, F4, and F3, and their influence on the classifier could explain why there is a little bit of doubt about the correctness of the C2 class assigned and the notable positive variables are F8, F5, F1, and F7.

{'C2': '89.16%', 'C3': '9.0%', 'C1': '1.84%'}

{'F9': 'IsActiveMember', 'F8': 'Age', 'F7': 'Geography', 'F10': 'NumOfProducts', 'F4': 'Gender', 'F5': 'Tenure', 'F3': 'CreditScore', 'F1': 'Balance', 'F6': 'EstimatedSalary', 'F2': 'HasCrCard'}

{'F9': 'F9', 'F4': 'F8', 'F2': 'F7', 'F7': 'F10', 'F3': 'F4', 'F5': 'F5', 'F1': 'F3', 'F6': 'F1', 'F10': 'F6', 'F8': 'F2'}

{'C1': 'C2', 'C2': 'C3', 'C3': 'C1'}

Stay

{'C2': 'Stay', 'C3': 'Leave', 'C1': 'Other'}

GradientBoostingClassifier

C2

Basketball Players Career Length Prediction

The case is labelled as C2 by the model but looking at the predicted probabilities across the different classes, there is a 33.63% chance that the label could be C1. To explain the above prediction conclusion, the analysis revealed that the majority of the features have negative influences or attributions, pushing the prediction away from C2 in favour of C1. The negative features include F12, F17, F6, F18, and F2 and the values of these features are ranked higher than any of the positive features. Shifting the prediction in the direction of C2 are the positive features F1, F13, F16, and F3. The analysis also revealed that the values of F19, F14, and F9 are less relevant to the prediction for the case under consideration.

{'C1': '33.63%', 'C2': '66.37%'}

{'F12': 'GamesPlayed', 'F17': 'OffensiveRebounds', 'F6': 'FieldGoalPercent', 'F18': 'FreeThrowPercent', 'F2': '3PointPercent', 'F1': '3PointAttempt', 'F8': 'FieldGoalsMade', 'F15': 'Blocks', 'F5': 'DefensiveRebounds', 'F4': 'Turnovers', 'F13': 'Rebounds', 'F7': 'FreeThrowAttempt', 'F16': 'MinutesPlayed', 'F10': 'Assists', 'F11': 'FieldGoalsAttempt', 'F3': '3PointMade', 'F19': 'PointsPerGame', 'F14': 'FreeThrowMade', 'F9': 'Steals'}

{'F1': 'F12', 'F13': 'F17', 'F6': 'F6', 'F12': 'F18', 'F9': 'F2', 'F8': 'F1', 'F4': 'F8', 'F18': 'F15', 'F14': 'F5', 'F19': 'F4', 'F15': 'F13', 'F11': 'F7', 'F2': 'F16', 'F16': 'F10', 'F5': 'F11', 'F7': 'F3', 'F3': 'F19', 'F10': 'F14', 'F17': 'F9'}

{'C1': 'C1', 'C2': 'C2'}

Less than 5

{'C1': 'More than 5', 'C2': 'Less than 5'}

BernoulliNB

C1

Customer Churn Modelling

The most likely label chosen by the model in this case is C1. The decision above is based on the prediction probabilities for the two possible labels, C1 and C2, which are 94.25% and 5.75%, respectively. The following variables can be ranked from most important to least important based on their contribution to the model when it comes to this instance: F7, F6, F1, F5, F10, F2, F3, F8, F4, and F9. F6 and F7 turned out to be the most important positive variables, supporting the model towards assigning the class C1. The least positive variables are F8 and F3, which have less effect on the model. In fact, most of the input features have negative contributions towards the assignment of class C1, leading to a decision change in favour of the other label, C2. The most negative variables are F10, F1, and F5, and the least negative are F4 and F9.

{'C1': '94.25%', 'C2': '5.75%'}

{'F7': 'IsActiveMember', 'F6': 'NumOfProducts', 'F1': 'Gender', 'F5': 'Geography', 'F10': 'Age', 'F2': 'CreditScore', 'F3': 'EstimatedSalary', 'F8': 'Balance', 'F4': 'HasCrCard', 'F9': 'Tenure'}

{'F9': 'F7', 'F7': 'F6', 'F3': 'F1', 'F2': 'F5', 'F4': 'F10', 'F1': 'F2', 'F10': 'F3', 'F6': 'F8', 'F8': 'F4', 'F5': 'F9'}

{'C1': 'C1', 'C2': 'C2'}

Stay

{'C1': 'Stay', 'C2': 'Leave'}

BernoulliNB

C2

Personal Loan Modelling

As per the classification algorithm employed, the most probable label for the data under consideration is C2 since the chances of C1 is very slim and negligible. The main driver behind the labelling decision above is F9. The features with moderate influence are F1, F7, F3, F4, F6, and F5, while those with very small or marginal impact are F2 and F8. The direction of influence of the input features could be used to explain why the algorithm is very confident here. Most of the features have a positive impact, increasing or improving the chances of C2 being the correct label and the feature with a significantly higher contribution, F9, is a positive feature which when coupled with other positives F7, F3, F5, and F6 encourages the prediction or assignment of the C2 label. Furthermore, aside from F1 and F4, the other two negative features, F2 and F8, are shown to have a significantly lower impact on the algorithm and the very marginal doubt in the decision can be attributed to the influence of the negative features.

{'C2': '99.99%', 'C1': '0.01%'}

{'F9': 'CD Account', 'F1': 'Income', 'F7': 'CCAvg', 'F3': 'Securities Account', 'F4': 'Education', 'F6': 'Family', 'F5': 'Mortgage', 'F2': 'Age', 'F8': 'Extra_service'}

{'F8': 'F9', 'F2': 'F1', 'F4': 'F7', 'F7': 'F3', 'F5': 'F4', 'F3': 'F6', 'F6': 'F5', 'F1': 'F2', 'F9': 'F8'}

{'C2': 'C2', 'C1': 'C1'}

Reject

{'C2': 'Reject', 'C1': 'Accept'}

MLPClassifier

C2

Vehicle Insurance Claims

The ML algorithm classifies the provided data or case as C2 with a likelihood of 80.70%, hinting that the likelihood of C1 being the correct label is only 19.30%. This classification decision above is mainly based on the influence or contributions of the input features. The most relevant features driving the classification algorithm to arrive at the above decision are F27, F23, F7, F1, F17, F30, and F3. On the other side, not all of the input features are considered relevant when deciding the appropriate label for the given data instance, and these irrelevant features include F2, F19, F25, F10, and F16. Among the top influential features, F17, F30, and F3 are regarded as negative features since their contributions push the algorithm's decision towards the less likely class, C1, although F27, F23, F7, F6, and F1 have positive contributions, increasing the probability that C2 is the right label here.

{'C1': '19.30%', 'C2': '80.70%'}

{'F27': 'incident_severity', 'F23': 'insured_relationship', 'F1': 'authorities_contacted', 'F7': 'vehicle_claim', 'F30': 'umbrella_limit', 'F3': 'insured_hobbies', 'F17': 'incident_type', 'F6': 'policy_deductable', 'F12': 'auto_make', 'F20': 'number_of_vehicles_involved', 'F13': 'insured_occupation', 'F8': 'property_damage', 'F5': 'incident_state', 'F15': 'auto_year', 'F22': 'capital-loss', 'F11': 'policy_csl', 'F28': 'collision_type', 'F24': 'capital-gains', 'F31': 'property_claim', 'F14': 'incident_hour_of_the_day', 'F2': 'police_report_available', 'F19': 'policy_annual_premium', 'F25': 'incident_city', 'F16': 'insured_zip', 'F10': 'bodily_injuries', 'F4': 'injury_claim', 'F9': 'witnesses', 'F26': 'total_claim_amount', 'F18': 'insured_education_level', 'F32': 'insured_sex', 'F29': 'policy_state', 'F21': 'age', 'F33': 'months_as_customer'}

{'F27': 'F27', 'F24': 'F23', 'F28': 'F1', 'F16': 'F7', 'F5': 'F30', 'F23': 'F3', 'F25': 'F17', 'F3': 'F6', 'F33': 'F12', 'F10': 'F20', 'F22': 'F13', 'F31': 'F8', 'F29': 'F5', 'F17': 'F15', 'F8': 'F22', 'F19': 'F11', 'F26': 'F28', 'F7': 'F24', 'F15': 'F31', 'F9': 'F14', 'F32': 'F2', 'F4': 'F19', 'F30': 'F25', 'F6': 'F16', 'F11': 'F10', 'F14': 'F4', 'F12': 'F9', 'F13': 'F26', 'F21': 'F18', 'F20': 'F32', 'F18': 'F29', 'F2': 'F21', 'F1': 'F33'}

{'C1': 'C1', 'C2': 'C2'}

Fraud

{'C1': 'Not Fraud', 'C2': 'Fraud'}

SVC

C1

Broadband Sevice Signup

The predicted probability of class C2 is 12.81% and that of class C1 is 87.19%. Therefore, the label chosen by the model is C1, which is the most probable class. The top two features with significant influence on the prediction verdict above are F4 and F27. These features have positive attributions, shifting the decision higher in support of label C1. Other positive features are F12, F41, F8, and F28. Decreasing the likelihood of the assigned label are the negative features such as F23, F29, F24, and F40. Finally, the values of features such as F39, F1, F37, F7, F33, and F31 are considered irrelevant to the prediction decision above.

{'C2': '12.81%', 'C1': '87.19%'}

{'F4': 'X38', 'F27': 'X32', 'F23': 'X22', 'F12': 'X35', 'F41': 'X25', 'F29': 'X16', 'F24': 'X12', 'F40': 'X31', 'F15': 'X3', 'F36': 'X9', 'F26': 'X1', 'F22': 'X19', 'F6': 'X4', 'F8': 'X2', 'F16': 'X29', 'F30': 'X42', 'F28': 'X36', 'F34': 'X21', 'F20': 'X40', 'F32': 'X10', 'F39': 'X33', 'F1': 'X5', 'F37': 'X6', 'F7': 'X41', 'F31': 'X39', 'F33': 'X7', 'F25': 'X37', 'F9': 'X8', 'F5': 'X34', 'F21': 'X18', 'F17': 'X17', 'F18': 'X11', 'F38': 'X30', 'F42': 'X28', 'F35': 'X27', 'F14': 'X26', 'F19': 'X13', 'F10': 'X14', 'F13': 'X23', 'F3': 'X15', 'F2': 'X20', 'F11': 'X24'}

{'F35': 'F4', 'F29': 'F27', 'F20': 'F23', 'F32': 'F12', 'F23': 'F41', 'F14': 'F29', 'F10': 'F24', 'F28': 'F40', 'F2': 'F15', 'F7': 'F36', 'F40': 'F26', 'F17': 'F22', 'F3': 'F6', 'F1': 'F8', 'F42': 'F16', 'F38': 'F30', 'F33': 'F28', 'F19': 'F34', 'F37': 'F20', 'F8': 'F32', 'F30': 'F39', 'F41': 'F1', 'F4': 'F37', 'F39': 'F7', 'F36': 'F31', 'F5': 'F33', 'F34': 'F25', 'F6': 'F9', 'F31': 'F5', 'F16': 'F21', 'F15': 'F17', 'F9': 'F18', 'F27': 'F38', 'F26': 'F42', 'F25': 'F35', 'F24': 'F14', 'F11': 'F19', 'F12': 'F10', 'F21': 'F13', 'F13': 'F3', 'F18': 'F2', 'F22': 'F11'}

{'C2': 'C2', 'C1': 'C1'}

Yes

{'C2': 'No', 'C1': 'Yes'}

MLPClassifier

C2

Ethereum Fraud Detection

The C1 has a predicted probability of just 3.10% while that of the C2 is 96.90%, therefore, the most likely class selected by the classifier for the given data is C2. The relevant features contributing to this classification are mainly F27, F11, F36, F33, F10, F4, F23, F15, F2, F8, F21, F32, F7, F37, F1, F13, F28, F30, F17, and F38. As per the attribution analysis, F27 and F11 have a very strong joint positive contribution, increasing the classifier's response higher in favour of C2 than C1. In contrast, F36, F10, and F33 are the top negative features, degrading the classifier's response in favour of C1. Comparing the attributions of F27, F4, and F11 to those of the negative features mentioned above, it is not surprising that the classifier is quite confident that C2 is the most probable label here.

{'C1': '3.10%', 'C2': '96.90%'}

{'F27': 'Unique Received From Addresses', 'F11': ' ERC20 total Ether sent contract', 'F36': 'total ether received', 'F33': 'Sent tnx', 'F10': 'Number of Created Contracts', 'F4': ' ERC20 uniq rec token name', 'F23': ' ERC20 uniq rec contract addr', 'F15': 'max value received ', 'F2': 'total transactions (including tnx to create contract', 'F8': ' ERC20 uniq sent addr.1', 'F21': ' ERC20 uniq sent addr', 'F32': 'Received Tnx', 'F7': 'avg val received', 'F37': ' ERC20 uniq rec addr', 'F1': 'avg val sent', 'F13': 'min value received', 'F28': 'Unique Sent To Addresses', 'F30': ' ERC20 uniq sent token name', 'F17': 'Avg min between received tnx', 'F38': 'Time Diff between first and last (Mins)', 'F35': ' ERC20 min val rec', 'F12': ' ERC20 max val rec', 'F22': ' ERC20 min val sent', 'F34': ' ERC20 max val sent', 'F19': ' ERC20 avg val sent', 'F5': ' ERC20 avg val rec', 'F31': ' Total ERC20 tnxs', 'F18': ' ERC20 total ether sent', 'F16': ' ERC20 total Ether received', 'F24': 'total ether balance', 'F9': 'total ether sent contracts', 'F25': 'total Ether sent', 'F6': 'avg value sent to contract', 'F3': 'max val sent to contract', 'F26': 'min value sent to contract', 'F29': 'max val sent', 'F14': 'min val sent', 'F20': 'Avg min between sent tnx'}

{'F7': 'F27', 'F26': 'F11', 'F20': 'F36', 'F4': 'F33', 'F6': 'F10', 'F38': 'F4', 'F30': 'F23', 'F10': 'F15', 'F18': 'F2', 'F29': 'F8', 'F27': 'F21', 'F5': 'F32', 'F11': 'F7', 'F28': 'F37', 'F14': 'F1', 'F9': 'F13', 'F8': 'F28', 'F37': 'F30', 'F2': 'F17', 'F3': 'F38', 'F31': 'F35', 'F32': 'F12', 'F34': 'F22', 'F35': 'F34', 'F36': 'F19', 'F33': 'F5', 'F23': 'F31', 'F25': 'F18', 'F24': 'F16', 'F22': 'F24', 'F21': 'F9', 'F19': 'F25', 'F17': 'F6', 'F16': 'F3', 'F15': 'F26', 'F13': 'F29', 'F12': 'F14', 'F1': 'F20'}

{'C1': 'C1', 'C2': 'C2'}

Fraud

{'C1': 'Not Fraud', 'C2': 'Fraud'}

RandomForestClassifier

C2

Printer Sales

There is only a 17.0% chance that C1 is the correct label which implies that the most probable label for the given data or case is C2 given its predicted likelihood of 83.0%. The main influential features resulting in the classification conclusions above are F26, F2, and F16 whereas the remaining features have either a moderate or negligible influence on the classifier. When it comes to assigning a label to this case, the classifier likely ignored the values of F8, F13, F21, F10, F22, and F17 since their respective degrees of influence are very close to zero. Among the influential features, only F20, F23, F19, F4, F12, and F14 are considered negative features mainly due to the fact that their contributions towards the decision here only serve to decrease the likelihood that C2 is the correct label and it can be said that these features favour labelling the case as C1. The remaining features such as F26, F2, F16, F6, F1, F3, and F24, offer positive contributions, increasing the likelihood of the C2 class.

{'C2': '83.00%', 'C1': '17.00%'}

{'F26': 'X8', 'F2': 'X24', 'F16': 'X1', 'F6': 'X2', 'F1': 'X10', 'F3': 'X15', 'F20': 'X25', 'F24': 'X23', 'F5': 'X18', 'F23': 'X4', 'F19': 'X7', 'F9': 'X17', 'F7': 'X3', 'F11': 'X22', 'F4': 'X5', 'F12': 'X9', 'F14': 'X12', 'F25': 'X19', 'F18': 'X11', 'F15': 'X16', 'F8': 'X14', 'F13': 'X21', 'F21': 'X20', 'F10': 'X13', 'F22': 'X6', 'F17': 'X26'}

{'F8': 'F26', 'F24': 'F2', 'F1': 'F16', 'F2': 'F6', 'F10': 'F1', 'F15': 'F3', 'F25': 'F20', 'F23': 'F24', 'F18': 'F5', 'F4': 'F23', 'F7': 'F19', 'F17': 'F9', 'F3': 'F7', 'F22': 'F11', 'F5': 'F4', 'F9': 'F12', 'F12': 'F14', 'F19': 'F25', 'F11': 'F18', 'F16': 'F15', 'F14': 'F8', 'F21': 'F13', 'F20': 'F21', 'F13': 'F10', 'F6': 'F22', 'F26': 'F17'}

{'C2': 'C2', 'C1': 'C1'}

Less

{'C2': 'Less', 'C1': 'More'}

RandomForestClassifier

C2

Student Job Placement

The classification algorithm's decision on the true label for the given case is solely dependent on the information presented to it. Per the algorithm, the accurate label for the case under consideration is most likely C2, and the 12.47% possibility of C1 reflects only a minor uncertainty in the classification algorithm's certainty. The marginal doubt mentioned above can be blamed on the negative contributions of F8, F5, F6, F12, and F2, supporting the assignment of C1 instead of C2. Conversely, the positive contributions of F10, F4, F1, F3, F9, and F11 are shifting the algorithm's decision higher in favour of label C2, hence the high certainty of its correctness. Overall, F8 and F5 are the most influential negative features, whereas F10 and F4 are the most positive features. Also, F7 is shown to have a negligible influence on the classification decision with respect to the case here.

{'C1': '12.47%', 'C2': '87.53%'}

{'F8': 'workex', 'F5': 'specialisation', 'F10': 'hsc_p', 'F4': 'gender', 'F1': 'mba_p', 'F6': 'hsc_s', 'F3': 'ssc_p', 'F9': 'etest_p', 'F12': 'ssc_b', 'F11': 'hsc_b', 'F2': 'degree_t', 'F7': 'degree_p'}

{'F11': 'F8', 'F12': 'F5', 'F2': 'F10', 'F6': 'F4', 'F5': 'F1', 'F9': 'F6', 'F1': 'F3', 'F4': 'F9', 'F7': 'F12', 'F8': 'F11', 'F10': 'F2', 'F3': 'F7'}

{'C2': 'C1', 'C1': 'C2'}

Placed

{'C1': 'Not Placed', 'C2': 'Placed'}

KNeighborsClassifier

C2

Printer Sales

The model indicates that the label for this case is likely C2, with an 83.33% chance that it is correct, implying that it is unlikely that C1 is the appropriate class. This predictive assertion is chiefly influenced by the values of the input variables F23, F24, and F10. While the F10 and F24 values positively control the model towards the prediction of C2, the F23 value biases the decision towards C1. However, the combined effect of F10 and F24 outweighs the contribution of F23. In addition, the variables F4, F6, and F18 also positively support the output predictions of the model. F21 has similar direction of contribution that of F23, further decreasing the odds of the C2 label. Unlike all the variables above, F11, F17, F8, F2, F12, and F19 are shown to have very little effect on model predictions with respect to the given case and we can say that their values receive very low consideration from the model.

{'C2': '83.33%', 'C1': '16.67%'}

{'F10': 'X24', 'F24': 'X1', 'F23': 'X4', 'F4': 'X10', 'F6': 'X2', 'F18': 'X8', 'F21': 'X17', 'F22': 'X7', 'F9': 'X21', 'F13': 'X18', 'F5': 'X6', 'F7': 'X11', 'F20': 'X22', 'F16': 'X25', 'F26': 'X5', 'F25': 'X19', 'F15': 'X15', 'F14': 'X23', 'F1': 'X16', 'F3': 'X3', 'F11': 'X14', 'F17': 'X20', 'F2': 'X13', 'F8': 'X12', 'F12': 'X9', 'F19': 'X26'}

{'F24': 'F10', 'F1': 'F24', 'F4': 'F23', 'F10': 'F4', 'F2': 'F6', 'F8': 'F18', 'F17': 'F21', 'F7': 'F22', 'F21': 'F9', 'F18': 'F13', 'F6': 'F5', 'F11': 'F7', 'F22': 'F20', 'F25': 'F16', 'F5': 'F26', 'F19': 'F25', 'F15': 'F15', 'F23': 'F14', 'F16': 'F1', 'F3': 'F3', 'F14': 'F11', 'F20': 'F17', 'F13': 'F2', 'F12': 'F8', 'F9': 'F12', 'F26': 'F19'}

{'C2': 'C2', 'C1': 'C1'}

Less

{'C2': 'Less', 'C1': 'More'}

LogisticRegression

C2

Hotel Satisfaction

The algorithm's forecast for the data instance under consideration is C2, and the decision's confidence level is about 91.36 percent. We can observe from the plot that the variables F8 and F3 are moving the prediction judgement towards the other label, C1. The F13, F6, F12, and F11, on the other hand, have values that have a favourable influence, pushing the data classification choice towards label C2. While F10 and F2 contradict the prediction, F4 and F9 have values that confirm the algorithm's prediction output verdict.

{'C2': '91.36%', 'C1': '8.64%'}

{'F8': 'Type of Travel', 'F3': 'Type Of Booking', 'F13': 'Hotel wifi service', 'F6': 'Common Room entertainment', 'F12': 'Stay comfort', 'F11': 'Other service', 'F10': 'Checkin\\/Checkout service', 'F4': 'Hotel location', 'F2': 'Food and drink', 'F9': 'Cleanliness', 'F14': 'Age', 'F15': 'Departure\\/Arrival convenience', 'F5': 'purpose_of_travel', 'F7': 'Ease of Online booking', 'F1': 'Gender'}

{'F3': 'F8', 'F4': 'F3', 'F6': 'F13', 'F12': 'F6', 'F11': 'F12', 'F14': 'F11', 'F13': 'F10', 'F9': 'F4', 'F10': 'F2', 'F15': 'F9', 'F5': 'F14', 'F7': 'F15', 'F2': 'F5', 'F8': 'F7', 'F1': 'F1'}

{'C2': 'C2', 'C1': 'C1'}

dissatisfied

{'C2': 'dissatisfied', 'C1': 'satisfied'}

SVC

C1

Vehicle Insurance Claims

The model classifies this case as C1 and it is noteworthy that there is, however, a 38.26% chance that the true label could be class C2. The uncertainty associated with the classification decision above is higher than expected, which could be attributed to the values of the different input features. The most influential feature is F18, which has a positive effect on the class C1 prediction by the model here. All other features are much less influential, with contributions from F30, F23, F17, and F27 shifting the prediction towards C2. Supporting the model in assigning the label choice, F15 is the next most influential feature. The impacts of the F9 and F12 are moderate, ranking seventh and eighth, respectively. Unfortunately, values of features such as F3, F24, F28, and F14 do not matter when determining the correct label in this instance.

{'C1': '61.74%', 'C2': '38.26%'}

{'F18': 'incident_severity', 'F30': 'insured_hobbies', 'F23': 'insured_occupation', 'F17': 'umbrella_limit', 'F27': 'policy_csl', 'F15': 'authorities_contacted', 'F9': 'insured_education_level', 'F12': 'collision_type', 'F22': 'months_as_customer', 'F25': 'vehicle_claim', 'F10': 'insured_relationship', 'F29': 'capital-gains', 'F21': 'auto_make', 'F7': 'injury_claim', 'F32': 'incident_city', 'F33': 'insured_sex', 'F5': 'number_of_vehicles_involved', 'F6': 'incident_hour_of_the_day', 'F8': 'age', 'F19': 'property_claim', 'F3': 'policy_annual_premium', 'F24': 'police_report_available', 'F28': 'property_damage', 'F14': 'incident_state', 'F4': 'policy_deductable', 'F26': 'capital-loss', 'F31': 'insured_zip', 'F13': 'incident_type', 'F11': 'bodily_injuries', 'F1': 'witnesses', 'F16': 'policy_state', 'F2': 'total_claim_amount', 'F20': 'auto_year'}

{'F27': 'F18', 'F23': 'F30', 'F22': 'F23', 'F5': 'F17', 'F19': 'F27', 'F28': 'F15', 'F21': 'F9', 'F26': 'F12', 'F1': 'F22', 'F16': 'F25', 'F24': 'F10', 'F7': 'F29', 'F33': 'F21', 'F14': 'F7', 'F30': 'F32', 'F20': 'F33', 'F10': 'F5', 'F9': 'F6', 'F2': 'F8', 'F15': 'F19', 'F4': 'F3', 'F32': 'F24', 'F31': 'F28', 'F29': 'F14', 'F3': 'F4', 'F8': 'F26', 'F6': 'F31', 'F25': 'F13', 'F11': 'F11', 'F12': 'F1', 'F18': 'F16', 'F13': 'F2', 'F17': 'F20'}

{'C1': 'C1', 'C2': 'C2'}

Fraud

{'C1': 'Fraud', 'C2': 'Not Fraud'}

SVM_linear

C2

Wine Quality Prediction

The classification or prediction algorithm indicates that the most probable label for the given data is C2 since there is only a 25.47% chance that C1 could be the correct label. The major factors resulting in the above decision are F1, F9, and F6, while the set of features with moderate influence are F8, F2, F5, and F4. The least vital features are shown to be F7, F11, F3, and F10. In conclusion, it is very surprising to see the uncertainty surrounding the classification here given that only F8 and F5 have a negative impact, driving the algorithm to label the data as C1. To be specific, the contributions of F8 and F5 result in a decrease in the likelihood of C2 being the right label, as indicated by the prediction probabilities across the two possible classes but the influence of these negatives are moderated by the major positive features which are F1, F9, and F6.

{'C1': '25.47%', 'C2': '74.53%'}

{'F1': 'sulphates', 'F9': 'volatile acidity', 'F6': 'total sulfur dioxide', 'F8': 'residual sugar', 'F2': 'alcohol', 'F5': 'free sulfur dioxide', 'F4': 'chlorides', 'F7': 'fixed acidity', 'F11': 'citric acid', 'F3': 'pH', 'F10': 'density'}

{'F10': 'F1', 'F2': 'F9', 'F7': 'F6', 'F4': 'F8', 'F11': 'F2', 'F6': 'F5', 'F5': 'F4', 'F1': 'F7', 'F3': 'F11', 'F9': 'F3', 'F8': 'F10'}

{'C2': 'C1', 'C1': 'C2'}

high quality

{'C1': 'low_quality', 'C2': 'high quality'}

RandomForestClassifier

C2

Flight Price-Range Classification

The classification verdict is as follows: the most probable label for this case is C2, and the classifier is certain that neither C1 nor C3 is the correct label. The main drivers for the above classification are F8, F6, and F9, all of which have a strong positive influence, pushing the classifier to choose C2. Other positive features pushing the classification further higher towards C2 include F2, F1, F3, and F10. Not all the input features support the assigned label and the negative features F12, F7, and F5 indicate that the most probable class for this case could different from the assigned label. However, considering the confidence level in the above classification, it is valid to conclude that the classifier paid little attention to the negative features, hence selecting class C2.

{'C2': '100.00%', 'C1': '0.00%', 'C3': '0.00%'}

{'F8': 'Airline', 'F6': 'Duration_hours', 'F9': 'Total_Stops', 'F2': 'Journey_month', 'F7': 'Source', 'F1': 'Destination', 'F3': 'Arrival_hour', 'F12': 'Journey_day', 'F5': 'Dep_minute', 'F10': 'Arrival_minute', 'F4': 'Duration_mins', 'F11': 'Dep_hour'}

{'F9': 'F8', 'F7': 'F6', 'F12': 'F9', 'F2': 'F2', 'F10': 'F7', 'F11': 'F1', 'F5': 'F3', 'F1': 'F12', 'F4': 'F5', 'F6': 'F10', 'F8': 'F4', 'F3': 'F11'}

{'C1': 'C2', 'C2': 'C1', 'C3': 'C3'}

Low

{'C2': 'Low', 'C1': 'Moderate', 'C3': 'High'}

RandomForestClassifier

C2

Ethereum Fraud Detection

The best choice of label for the given case is C2 according to the classification algorithm, since there is little to no chance that C1 is the right class. Not all the features are shown to contribute either positively or negatively towards the label assigned here. The influential features can be ranked according to the associated degree of impact on the algorithm's output as follows: F33, F7, F18, F36, F1, F12, F31, F21, F27, F8, F38, F20, F26, F30, F2, F34, F11, F25, F37, F15. On the other hand, the irrelevant features include F3, F6, and F17 since they have close to zero impact. Among the top influential ones, F33, F7, F18, F36, and F1, the input feature F18 is regarded as the most negative, dragging the verdict in a different direction, while the others have positive contributions, improving the likelihood that the choice of C2 is appropriate in this case. The features with moderate influence are F31, F12, F21 where F31 is identified as a positive feature, while F12 and F21 considered negative features. Since a large number of top features have positive contributions that increase the probability that C2 is the right label, it is not surprising that the algorithm is very confident about the correctness of the assigned label.

{'C1': '0.00%', 'C2': '100.00%'}

{'F33': ' ERC20 total Ether sent contract', 'F7': ' ERC20 min val rec', 'F18': 'total transactions (including tnx to create contract', 'F36': ' ERC20 max val rec', 'F1': ' Total ERC20 tnxs', 'F12': ' ERC20 uniq rec addr', 'F31': 'min val sent', 'F21': 'Time Diff between first and last (Mins)', 'F27': 'Sent tnx', 'F8': 'Avg min between received tnx', 'F38': 'min value received', 'F20': ' ERC20 total ether sent', 'F26': 'avg val sent', 'F30': 'max val sent', 'F2': 'Avg min between sent tnx', 'F34': 'Received Tnx', 'F11': ' ERC20 uniq sent token name', 'F25': 'Unique Sent To Addresses', 'F37': ' ERC20 uniq rec token name', 'F15': ' ERC20 uniq rec contract addr', 'F3': 'total Ether sent', 'F6': 'Number of Created Contracts', 'F17': ' ERC20 avg val sent', 'F16': ' ERC20 max val sent', 'F35': ' ERC20 min val sent', 'F24': ' ERC20 avg val rec', 'F32': 'Unique Received From Addresses', 'F29': 'max value received ', 'F22': ' ERC20 uniq sent addr.1', 'F10': 'total ether sent contracts', 'F19': 'avg val received', 'F14': ' ERC20 uniq sent addr', 'F5': 'min value sent to contract', 'F9': 'max val sent to contract', 'F23': ' ERC20 total Ether received', 'F13': 'avg value sent to contract', 'F28': 'total ether balance', 'F4': 'total ether received'}

{'F26': 'F33', 'F31': 'F7', 'F18': 'F18', 'F32': 'F36', 'F23': 'F1', 'F28': 'F12', 'F12': 'F31', 'F3': 'F21', 'F4': 'F27', 'F2': 'F8', 'F9': 'F38', 'F25': 'F20', 'F14': 'F26', 'F13': 'F30', 'F1': 'F2', 'F5': 'F34', 'F37': 'F11', 'F8': 'F25', 'F38': 'F37', 'F30': 'F15', 'F19': 'F3', 'F6': 'F6', 'F36': 'F17', 'F35': 'F16', 'F34': 'F35', 'F33': 'F24', 'F7': 'F32', 'F10': 'F29', 'F29': 'F22', 'F21': 'F10', 'F11': 'F19', 'F27': 'F14', 'F15': 'F5', 'F16': 'F9', 'F24': 'F23', 'F17': 'F13', 'F22': 'F28', 'F20': 'F4'}

{'C1': 'C1', 'C2': 'C2'}

Fraud

{'C1': 'Not Fraud', 'C2': 'Fraud'}

LogisticRegression

C1

Student Job Placement

For the given case, the prediction decision is as follows: The probability of C2 being the correct label is only 18.57%, the probability of C1 is 81.43% making it the most probable label for the case here. The certainty of the prediction can be attributed to the influence of variables such as F5, F1, F7, F12, and F6. The least relevant variables considered to arrive at the classification verdict are F10, F3, F4, and F2. F9, F11, and F8 have moderate contributions to the classification here. The attribution analysis performed indicates that F12, F6, F11, F8, F3, and F2 are the negative variables, decreasing the likelihood of C1 in favour of labelling the given case as C2. The variables F5, F1, and F7 have the highest positive influence, which increases the odds of label C1 being the correct label.

{'C2': '18.57%', 'C1': '81.43%'}

{'F5': 'mba_p', 'F1': 'gender', 'F7': 'degree_t', 'F12': 'specialisation', 'F6': 'workex', 'F9': 'hsc_s', 'F11': 'hsc_p', 'F8': 'ssc_p', 'F10': 'ssc_b', 'F3': 'etest_p', 'F4': 'hsc_b', 'F2': 'degree_p'}

{'F5': 'F5', 'F6': 'F1', 'F10': 'F7', 'F12': 'F12', 'F11': 'F6', 'F9': 'F9', 'F2': 'F11', 'F1': 'F8', 'F7': 'F10', 'F4': 'F3', 'F8': 'F4', 'F3': 'F2'}

{'C2': 'C2', 'C1': 'C1'}

Placed

{'C2': 'Not Placed', 'C1': 'Placed'}

SGDClassifier

C2

Flight Price-Range Classification

The output decision of the classifier with respect to the given case is: C2 is the most probable label, followed by C3 and C1. To be specific, the predicted likelihood across the classes are as follows: 86.54% for C2, 13.46% for C3, and finally a 0.0% probability with respect to C1. The moderately high classification confidence could largely be due to the impact of certain input features supplied to the classifier. F3, F4, F9, F1, and F6 are the top-ranked variables whereas the least ranked are F8, F2, F7, F5, F10, F11, and F12. The marginal uncertainty in the classification verdict is due to the negative attributions of F4, F6, F7, F10, and F12 which prefer labelling the case differently. In conclusion, we can see that F3, F9, F1, F2, and F8 are among the positive variables pushing the classification in favour of C2.

{'C2': '86.54%', 'C3': '13.46%', 'C1': '0.00%'}

{'F3': 'Airline', 'F4': 'Total_Stops', 'F9': 'Source', 'F1': 'Journey_month', 'F6': 'Arrival_minute', 'F8': 'Journey_day', 'F2': 'Duration_hours', 'F7': 'Dep_hour', 'F5': 'Destination', 'F10': 'Arrival_hour', 'F11': 'Dep_minute', 'F12': 'Duration_mins'}

{'F9': 'F3', 'F12': 'F4', 'F10': 'F9', 'F2': 'F1', 'F6': 'F6', 'F1': 'F8', 'F7': 'F2', 'F3': 'F7', 'F11': 'F5', 'F5': 'F10', 'F4': 'F11', 'F8': 'F12'}

{'C1': 'C2', 'C2': 'C3', 'C3': 'C1'}

Low

{'C2': 'Low', 'C3': 'Moderate', 'C1': 'High'}

LogisticRegression

C2

Airline Passenger Satisfaction

C2 is the predicted label assigned to this case or instance. This is based on the fact that there is only a 0.68% chance that C1 is the correct label. The most relevant variables that increase the prediction's probability are F12, F2, F19, and F17. Conversely, F13 is the only important feature driving the classification decision in the direction of C1. Other negative features include F21, F7, F8, and F18. Other positive features increasing the chances of the C2 prediction are F15, F22, and F5. Unlike F12, F2, F19, and F17, these positive variables have moderate contributions to the model's decision. The least ranked among all the relevant features are F16, F1, F11, and F20, with lower attributions to the C2 prediction, however, F10 and F4 are shown to have no impact when determining the correct label for the case under consideration.

{'C2': '99.32%', 'C1': '0.68%'}

{'F12': 'Type of Travel', 'F13': 'Customer Type', 'F2': 'Inflight entertainment', 'F17': 'Inflight wifi service', 'F19': 'Departure\\/Arrival time convenient', 'F21': 'Gate location', 'F7': 'Arrival Delay in Minutes', 'F8': 'Seat comfort', 'F18': 'Online boarding', 'F15': 'Ease of Online booking', 'F22': 'Class', 'F5': 'Age', 'F6': 'On-board service', 'F3': 'Cleanliness', 'F9': 'Checkin service', 'F14': 'Inflight service', 'F16': 'Food and drink', 'F1': 'Departure Delay in Minutes', 'F11': 'Baggage handling', 'F20': 'Gender', 'F10': 'Flight Distance', 'F4': 'Leg room service'}

{'F4': 'F12', 'F2': 'F13', 'F14': 'F2', 'F7': 'F17', 'F8': 'F19', 'F10': 'F21', 'F22': 'F7', 'F13': 'F8', 'F12': 'F18', 'F9': 'F15', 'F5': 'F22', 'F3': 'F5', 'F15': 'F6', 'F20': 'F3', 'F18': 'F9', 'F19': 'F14', 'F11': 'F16', 'F21': 'F1', 'F17': 'F11', 'F1': 'F20', 'F6': 'F10', 'F16': 'F4'}

{'C2': 'C2', 'C1': 'C1'}

neutral or dissatisfied

{'C2': 'neutral or dissatisfied', 'C1': 'satisfied'}

RandomForestClassifier

C2

Personal Loan Modelling

The following classification decisions are largely based on the factors or attributes of this particular case. The class label, in this case, is projected to be C2 out of the potential classes, which is 97.50% likely. The next possible label is C1, which has an approximate probability of 2.50%. The confidence level with respect to this classification is very high, and the features with the most contributions are F5, F9, and F4. However, F6, F1, and F7 are shown to be the least relevant features. The attribution analysis shows that the only positive features whose contributions favour labelling the case as C2 are F9, F4, F8, and F3. However, the negative attributions of F5, F2, F1, F6, and F7 also indicate that perhaps C1 could be the true label. Judging based on the confidence level coupled with the attributions, it can be concluded that the values of the positive features F9, F4, F8, and F3 are good enough to steer the classification in the direction of C2, but the strong negative attribution of F5 casts about 2.50% of doubt on the decision.

{'C2': '97.50%', 'C1': '2.50%'}

{'F5': 'Income', 'F9': 'CD Account', 'F4': 'Education', 'F2': 'Securities Account', 'F8': 'CCAvg', 'F3': 'Family', 'F1': 'Extra_service', 'F6': 'Age', 'F7': 'Mortgage'}

{'F2': 'F5', 'F8': 'F9', 'F5': 'F4', 'F7': 'F2', 'F4': 'F8', 'F3': 'F3', 'F9': 'F1', 'F1': 'F6', 'F6': 'F7'}

{'C1': 'C2', 'C2': 'C1'}

Reject

{'C2': 'Reject', 'C1': 'Accept'}

DNN

C1

Ethereum Fraud Detection

The prediction probabilities for classes C2 and C1, respectively, are 15.35% and 84.65%. Based on the aforementioned, C1 is the most likely class label for the presented data instance, and according to the attribution analysis, the various input variables had varying degrees of impact on the model's classification judgement. F3, F38, F2, F4, F32, F24, and F11 are the most influential factors, whereas F17, F29, F33, F25, F35, and F16 have the least impact. The subsequent analysis will concentrate on the most relevant factors influencing the label selection in this case. Looking at the attributions of the input features, only F3 and F38 exhibit negative contributions among the top influential features, F3, F38, F2, F4, and F11, lowering the chance that C1 is the right label, and they strongly favour labelling the instance as C2 instead. Positive variables such as F2, F4, and F11 influence the classification choice in favour of C1. The remaining variables, including F32, F24, and F10, have a moderate to low impact. In essence, the marginal uncertainty in this decision is mostly owing to the negative impacts of F3, F38, F31, and F14, while the positive contributions of F2, F4, F10, F32, F24, and F11 push the decision much closer to C1.

{'C2': '15.35%', 'C1': '84.65%'}

{'F3': ' ERC20 uniq rec contract addr', 'F38': ' ERC20 uniq rec token name', 'F2': 'min value received', 'F4': 'Time Diff between first and last (Mins)', 'F11': 'avg val sent', 'F32': ' ERC20 uniq sent token name', 'F24': 'Sent tnx', 'F10': 'Avg min between received tnx', 'F31': 'Unique Received From Addresses', 'F14': ' ERC20 uniq rec addr', 'F13': 'total transactions (including tnx to create contract', 'F5': 'Avg min between sent tnx', 'F22': ' ERC20 uniq sent addr.1', 'F23': 'avg val received', 'F20': 'Unique Sent To Addresses', 'F21': 'max value received ', 'F28': 'max val sent', 'F36': 'min val sent', 'F15': 'Number of Created Contracts', 'F1': 'total ether received', 'F9': ' ERC20 uniq sent addr', 'F12': ' ERC20 total Ether received', 'F18': 'Received Tnx', 'F34': ' ERC20 avg val sent', 'F7': 'total Ether sent', 'F37': ' ERC20 min val sent', 'F19': 'max val sent to contract', 'F6': 'total ether balance', 'F8': ' ERC20 max val sent', 'F30': ' Total ERC20 tnxs', 'F26': ' ERC20 total ether sent', 'F27': ' ERC20 avg val rec', 'F33': 'avg value sent to contract', 'F29': ' ERC20 min val rec', 'F17': ' ERC20 max val rec', 'F16': ' ERC20 total Ether sent contract', 'F35': 'min value sent to contract', 'F25': 'total ether sent contracts'}

{'F30': 'F3', 'F38': 'F38', 'F9': 'F2', 'F3': 'F4', 'F14': 'F11', 'F37': 'F32', 'F4': 'F24', 'F2': 'F10', 'F7': 'F31', 'F28': 'F14', 'F18': 'F13', 'F1': 'F5', 'F29': 'F22', 'F11': 'F23', 'F8': 'F20', 'F10': 'F21', 'F13': 'F28', 'F12': 'F36', 'F6': 'F15', 'F20': 'F1', 'F27': 'F9', 'F24': 'F12', 'F5': 'F18', 'F36': 'F34', 'F19': 'F7', 'F34': 'F37', 'F16': 'F19', 'F22': 'F6', 'F35': 'F8', 'F23': 'F30', 'F25': 'F26', 'F33': 'F27', 'F17': 'F33', 'F31': 'F29', 'F32': 'F17', 'F26': 'F16', 'F15': 'F35', 'F21': 'F25'}

{'C2': 'C2', 'C1': 'C1'}

Fraud

{'C2': 'Not Fraud', 'C1': 'Fraud'}

RandomForestClassifier

C1

Company Bankruptcy Prediction

The model assigns the class C1 with near perfect certainty or confidence level since the predicted likelihood of C2 is only 1.0%. F88, F46, F44, F54, and F50 have the greatest cumulative beneficial influence on the model's choice to create C1. F56 also had a significant influence, but it shifted the choice away from C1. Furthermore, F19 and F10 had a modest influence on C1 decision making, which was still bigger than features F14 and F47, which had a moderate impact and contributed to C2 class prediction. Furthermore, F10, F74, and F16 have minimal positive impact on the final result, further increasing the chances of C1 being the appropriate label for the given case. However, a number of input features, notably F26, F7, F89, and F42, appear to be less essential to predictions here. All in all, the very high confidence level could easily be explained away by considering the fact that the joint influence of the positive variables such as F88, F46, F44, F54, and F50 far outshines the joint contribution of the negative variables such as F56, F14, F47, and F39.

{'C1': '99.00%', 'C2': '1.00%'}

{'F88': " Net Income to Stockholder's Equity", 'F44': ' Continuous interest rate (after tax)', 'F54': ' ROA(C) before interest and depreciation before interest', 'F56': ' Borrowing dependency', 'F46': ' Cash Flow Per Share', 'F50': ' Net worth\\/Assets', 'F19': ' Total income\\/Total expense', 'F10': ' Persistent EPS in the Last Four Seasons', 'F74': ' Retained Earnings to Total Assets', 'F16': ' Net Value Per Share (B)', 'F1': ' Cash Flow to Equity', 'F34': ' Net Value Per Share (A)', 'F48': ' Degree of Financial Leverage (DFL)', 'F14': ' Per Share Net profit before tax (Yuan ¥)', 'F47': ' Revenue Per Share (Yuan ¥)', 'F39': ' Inventory Turnover Rate (times)', 'F40': ' Net profit before tax\\/Paid-in capital', 'F75': ' Equity to Long-term Liability', 'F82': ' Operating profit\\/Paid-in capital', 'F13': ' Cash Turnover Rate', 'F26': ' Operating Funds to Liability', 'F7': ' Contingent liabilities\\/Net worth', 'F89': ' Working Capital to Total Assets', 'F42': ' Liability to Equity', 'F57': ' Current Liability to Liability', 'F41': ' Operating Gross Margin', 'F45': ' Operating Profit Per Share (Yuan ¥)', 'F33': ' Long-term Liability to Current Assets', 'F70': ' Current Asset Turnover Rate', 'F71': ' Interest Coverage Ratio (Interest expense to EBIT)', 'F81': ' Equity to Liability', 'F32': ' Operating Profit Rate', 'F18': ' Current Liability to Equity', 'F31': ' No-credit Interval', 'F77': ' Net Worth Turnover Rate (times)', 'F63': ' Working Capital\\/Equity', 'F73': ' Quick Assets\\/Current Liability', 'F49': ' Inventory and accounts receivable\\/Net value', 'F24': ' Current Liability to Current Assets', 'F4': ' Working capitcal Turnover Rate', 'F3': ' Fixed Assets to Assets', 'F66': ' Continuous Net Profit Growth Rate', 'F67': ' Cash Reinvestment %', 'F37': ' CFO to Assets', 'F25': ' Total Asset Turnover', 'F36': ' After-tax net Interest Rate', 'F60': ' After-tax Net Profit Growth Rate', 'F53': ' Tax rate (A)', 'F22': ' Current Ratio', 'F43': ' Realized Sales Gross Margin', 'F86': ' Net Value Per Share (C)', 'F35': ' Regular Net Profit Growth Rate', 'F68': ' Interest-bearing debt interest rate', 'F38': ' Debt ratio %', 'F90': ' Long-term fund suitability ratio (A)', 'F64': ' Net Value Growth Rate', 'F58': ' Total Asset Growth Rate', 'F12': ' Fixed Assets Turnover Frequency', 'F9': ' Inventory\\/Current Liability', 'F21': ' Allocation rate per person', 'F2': ' Operating Expense Rate', 'F79': ' Operating profit per person', 'F76': ' Net Income to Total Assets', 'F92': ' Interest Expense Ratio', 'F61': ' Cash\\/Total Assets', 'F8': ' ROA(B) before interest and depreciation after tax', 'F87': ' Inventory\\/Working Capital', 'F17': ' Total assets to GNP price', 'F59': ' Total debt\\/Total net worth', 'F5': ' Quick Ratio', 'F23': ' Revenue per person', 'F72': ' Non-industry income and expenditure\\/revenue', 'F65': ' Cash Flow to Sales', 'F20': ' ROA(A) before interest and % after tax', 'F93': ' Current Liabilities\\/Liability', 'F27': ' Operating Profit Growth Rate', 'F51': ' Cash Flow to Liability', 'F6': ' Cash Flow to Total Assets', 'F15': ' Pre-tax net Interest Rate', 'F85': ' Accounts Receivable Turnover', 'F30': ' Current Liability to Assets', 'F78': ' Quick Assets\\/Total Assets', 'F69': ' Total expense\\/Assets', 'F84': ' Average Collection Days', 'F11': ' Research and development expense rate', 'F83': ' Current Assets\\/Total Assets', 'F91': ' Current Liabilities\\/Equity', 'F28': ' Realized Sales Gross Profit Growth Rate', 'F29': ' Cash flow rate', 'F52': ' Total Asset Return Growth Rate Ratio', 'F80': ' Quick Asset Turnover Rate', 'F55': ' Cash\\/Current Liability', 'F62': ' Gross Profit to Sales'}

{'F59': 'F88', 'F12': 'F44', 'F29': 'F54', 'F3': 'F56', 'F65': 'F46', 'F84': 'F50', 'F57': 'F19', 'F8': 'F10', 'F10': 'F74', 'F27': 'F16', 'F53': 'F1', 'F42': 'F34', 'F35': 'F48', 'F78': 'F14', 'F31': 'F47', 'F18': 'F39', 'F72': 'F40', 'F23': 'F75', 'F89': 'F82', 'F34': 'F13', 'F87': 'F26', 'F64': 'F7', 'F67': 'F89', 'F66': 'F42', 'F90': 'F57', 'F62': 'F41', 'F63': 'F45', 'F69': 'F33', 'F61': 'F70', 'F60': 'F71', 'F91': 'F81', 'F58': 'F32', 'F92': 'F18', 'F56': 'F31', 'F55': 'F77', 'F68': 'F63', 'F71': 'F73', 'F70': 'F49', 'F86': 'F24', 'F73': 'F4', 'F74': 'F3', 'F54': 'F66', 'F75': 'F67', 'F76': 'F37', 'F77': 'F25', 'F79': 'F36', 'F80': 'F60', 'F81': 'F53', 'F82': 'F22', 'F83': 'F43', 'F88': 'F86', 'F85': 'F35', 'F1': 'F68', 'F47': 'F38', 'F52': 'F90', 'F15': 'F64', 'F24': 'F58', 'F22': 'F12', 'F21': 'F9', 'F20': 'F21', 'F19': 'F2', 'F17': 'F79', 'F16': 'F76', 'F14': 'F92', 'F26': 'F61', 'F13': 'F8', 'F11': 'F87', 'F9': 'F17', 'F7': 'F59', 'F6': 'F5', 'F5': 'F23', 'F4': 'F72', 'F25': 'F65', 'F28': 'F20', 'F51': 'F93', 'F43': 'F27', 'F50': 'F51', 'F49': 'F6', 'F48': 'F15', 'F2': 'F85', 'F46': 'F30', 'F45': 'F78', 'F44': 'F69', 'F41': 'F84', 'F30': 'F11', 'F40': 'F83', 'F39': 'F91', 'F38': 'F28', 'F37': 'F29', 'F36': 'F52', 'F33': 'F80', 'F32': 'F55', 'F93': 'F62'}

{'C1': 'C1', 'C2': 'C2'}

No

{'C1': 'No', 'C2': 'Yes'}

RandomForestClassifier

C2

House Price Classification

Between the two classes, the model labelled this case as C2 with a likelihood of about 97.0% since there is only a marginal chance that it belongs to label C1. The most relevant features influencing this decision are F5, F3, F12, and F1. In this case, F5, F3, and F1 have a considerable positive influence on the prediction of C2. In contrast, the values of F12 and F9 throw a bit of doubt on the C2 prediction. However, compared to F5, F3, and F1, this shift is very small. Finally, there are some attributes with limited impact on the prediction of C2 and these are F11, F4, F2, F13, F10, and F6 since their values are less important to the model in terms of determining the label for this case.

{'C1': '3.00%', 'C2': '97.00%'}

{'F5': 'LSTAT', 'F3': 'RM', 'F1': 'AGE', 'F12': 'TAX', 'F8': 'PTRATIO', 'F9': 'DIS', 'F7': 'CRIM', 'F11': 'RAD', 'F4': 'B', 'F2': 'NOX', 'F13': 'ZN', 'F10': 'INDUS', 'F6': 'CHAS'}

{'F13': 'F5', 'F6': 'F3', 'F7': 'F1', 'F10': 'F12', 'F11': 'F8', 'F8': 'F9', 'F1': 'F7', 'F9': 'F11', 'F12': 'F4', 'F5': 'F2', 'F2': 'F13', 'F3': 'F10', 'F4': 'F6'}

{'C2': 'C1', 'C1': 'C2'}

High

{'C1': 'Low', 'C2': 'High'}

LogisticRegression

C2

Concrete Strength Classification

Probably C2 is the right label for this case since the probability of the alternative label, C3 and C1, are only 1.03% and 0.0%. The order of importance of the features for the above classification verdict is F5, F8, F7, F3, F6, F2, F1, and F4. Analysis conducted shows that only the features F8, F6, and F2 have negative contributions, hence reducing the probability of assigning label C2 to the given case. Positive features that increase the likelihood that C2 is the valid label are F5, F7, F3, F1, and F4. The co-attribution of the positive variables is stronger than that of the negative ones, so it is not surprising that we see the level of confidence associated with the prediction of class C2.

{'C3': '1.03%', 'C2': '98.97%', 'C1': '0.0%'}

{'F5': 'cement', 'F8': 'age_days', 'F7': 'water', 'F3': 'superplasticizer', 'F6': 'fineaggregate', 'F2': 'flyash', 'F1': 'slag', 'F4': 'coarseaggregate'}

{'F1': 'F5', 'F8': 'F8', 'F4': 'F7', 'F5': 'F3', 'F7': 'F6', 'F3': 'F2', 'F2': 'F1', 'F6': 'F4'}

{'C3': 'C3', 'C1': 'C2', 'C2': 'C1'}

Strong

{'C3': 'Weak', 'C2': 'Strong', 'C1': 'Other'}

DNN

C2

Credit Card Fraud Classification

The model labels the given data as C2 since it has a higher predicted probability equal to 51.42% compared to that of C1 which is equal to 48.58%. The input variables with higher contributions to the above classification decision are F6, F15, F21, F20, and F9, while those with little influence are F17, F2, F19, F12, and F22. Positively supporting the choice of the label, in this case, are mainly F6, F15, F9, and F20. However, the main negative variables are F21, F18, and F10. Judging based on the degree of influence as well as the direction of influence of the variables, it is not surprising that the model is only 51.42% confident in the assigned label which is marginally above average.

{'C1': '48.58%', 'C2': '51.42%'}

{'F15': 'Z18', 'F6': 'Z14', 'F21': 'Time', 'F20': 'Z1', 'F9': 'Z19', 'F1': 'Z10', 'F23': 'Z4', 'F25': 'Z3', 'F26': 'Z12', 'F13': 'Z16', 'F18': 'Z7', 'F10': 'Z11', 'F4': 'Z9', 'F11': 'Z6', 'F28': 'Z23', 'F30': 'Z5', 'F24': 'Z17', 'F16': 'Z21', 'F3': 'Z24', 'F8': 'Z8', 'F14': 'Amount', 'F29': 'Z20', 'F5': 'Z27', 'F27': 'Z25', 'F7': 'Z13', 'F17': 'Z2', 'F2': 'Z22', 'F19': 'Z28', 'F12': 'Z26', 'F22': 'Z15'}

{'F19': 'F15', 'F15': 'F6', 'F1': 'F21', 'F2': 'F20', 'F20': 'F9', 'F11': 'F1', 'F5': 'F23', 'F4': 'F25', 'F13': 'F26', 'F17': 'F13', 'F8': 'F18', 'F12': 'F10', 'F10': 'F4', 'F7': 'F11', 'F24': 'F28', 'F6': 'F30', 'F18': 'F24', 'F22': 'F16', 'F25': 'F3', 'F9': 'F8', 'F30': 'F14', 'F21': 'F29', 'F28': 'F5', 'F26': 'F27', 'F14': 'F7', 'F3': 'F17', 'F23': 'F2', 'F29': 'F19', 'F27': 'F12', 'F16': 'F22'}

{'C1': 'C1', 'C2': 'C2'}

Fraud

{'C1': 'Not Fraud', 'C2': 'Fraud'}

SGDClassifier

C2

Company Bankruptcy Prediction

The following is the classification for the provided data: C2 is the most likely class label and C1 cannot possibly be the correct label given the likelihood is 0.0%. F14, F66, and F1 are the key variables that contributed to the classification choice. However, the classifier does not consider all features while making this conclusion, and these irrelevant features include F78, F83, F40, and F75. Revealed to have positive contributions to the prediction made here among the top features are F1, F80, F16, and F74, but all of the others, F14, F66, F34, F85, F82, F55, and F33, argue against labelling the present scenario as C1 and despite the fact that the bulk of relevant features are pointing in the opposite direction, the classifier is extremely certain that the proper label for the current scenario is C2, not C1.

{'C1': '0.00%', 'C2': '100.00%'}

{'F14': ' Liability to Equity', 'F66': ' Net worth\\/Assets', 'F1': ' Debt ratio %', 'F34': " Net Income to Stockholder's Equity", 'F33': ' Equity to Liability', 'F74': ' Realized Sales Gross Margin', 'F85': ' Net Value Per Share (A)', 'F82': ' Current Liability to Assets', 'F80': ' Current Liability to Equity', 'F16': ' Net Income to Total Assets', 'F55': ' Operating Profit Per Share (Yuan ¥)', 'F63': ' ROA(B) before interest and depreciation after tax', 'F30': ' Working Capital to Total Assets', 'F65': ' Persistent EPS in the Last Four Seasons', 'F68': ' Current Liabilities\\/Equity', 'F89': ' Total expense\\/Assets', 'F44': ' Net Value Per Share (C)', 'F19': ' Gross Profit to Sales', 'F9': ' Pre-tax net Interest Rate', 'F24': ' Cash\\/Current Liability', 'F78': ' Total assets to GNP price', 'F83': ' Working capitcal Turnover Rate', 'F40': ' Net profit before tax\\/Paid-in capital', 'F75': ' Quick Assets\\/Current Liability', 'F36': ' Inventory and accounts receivable\\/Net value', 'F81': ' Long-term Liability to Current Assets', 'F62': ' Working Capital\\/Equity', 'F58': ' Operating Expense Rate', 'F48': ' Cash Reinvestment %', 'F61': ' Retained Earnings to Total Assets', 'F27': ' Cash Flow Per Share', 'F86': ' Contingent liabilities\\/Net worth', 'F22': ' Inventory\\/Working Capital', 'F73': ' Operating Gross Margin', 'F25': ' Current Asset Turnover Rate', 'F92': ' Interest Coverage Ratio (Interest expense to EBIT)', 'F3': ' Fixed Assets to Assets', 'F90': ' CFO to Assets', 'F67': ' Operating Profit Rate', 'F41': ' Total Asset Turnover', 'F29': ' Borrowing dependency', 'F11': ' Non-industry income and expenditure\\/revenue', 'F60': ' Current Liability to Liability', 'F71': ' Operating profit\\/Paid-in capital', 'F20': ' Revenue per person', 'F21': ' Operating Funds to Liability', 'F56': ' Current Liability to Current Assets', 'F59': ' Regular Net Profit Growth Rate', 'F35': ' Quick Ratio', 'F38': ' Total debt\\/Total net worth', 'F87': ' Current Ratio', 'F5': ' Tax rate (A)', 'F88': ' After-tax Net Profit Growth Rate', 'F53': ' After-tax net Interest Rate', 'F4': ' Per Share Net profit before tax (Yuan ¥)', 'F17': ' Continuous interest rate (after tax)', 'F39': ' No-credit Interval', 'F45': ' Total income\\/Total expense', 'F26': ' Allocation rate per person', 'F64': ' Total Asset Return Growth Rate Ratio', 'F23': ' Degree of Financial Leverage (DFL)', 'F13': ' Cash Turnover Rate', 'F43': ' Quick Asset Turnover Rate', 'F79': ' Revenue Per Share (Yuan ¥)', 'F51': ' Research and development expense rate', 'F49': ' ROA(C) before interest and depreciation before interest', 'F7': ' ROA(A) before interest and % after tax', 'F72': ' Net Value Per Share (B)', 'F84': ' Cash\\/Total Assets', 'F32': ' Cash Flow to Sales', 'F54': ' Total Asset Growth Rate', 'F77': ' Equity to Long-term Liability', 'F6': ' Fixed Assets Turnover Frequency', 'F28': ' Inventory\\/Current Liability', 'F69': ' Cash flow rate', 'F50': ' Realized Sales Gross Profit Growth Rate', 'F91': ' Inventory Turnover Rate (times)', 'F10': ' Cash Flow to Total Assets', 'F70': ' Net Worth Turnover Rate (times)', 'F47': ' Continuous Net Profit Growth Rate', 'F46': ' Cash Flow to Equity', 'F31': ' Long-term fund suitability ratio (A)', 'F15': ' Current Liabilities\\/Liability', 'F18': ' Cash Flow to Liability', 'F52': ' Accounts Receivable Turnover', 'F12': ' Current Assets\\/Total Assets', 'F93': ' Interest Expense Ratio', 'F76': ' Quick Assets\\/Total Assets', 'F57': ' Net Value Growth Rate', 'F42': ' Operating Profit Growth Rate', 'F37': ' Operating profit per person', 'F8': ' Average Collection Days', 'F2': ' Interest-bearing debt interest rate'}

{'F66': 'F14', 'F84': 'F66', 'F47': 'F1', 'F59': 'F34', 'F91': 'F33', 'F83': 'F74', 'F42': 'F85', 'F46': 'F82', 'F92': 'F80', 'F16': 'F16', 'F63': 'F55', 'F13': 'F63', 'F67': 'F30', 'F8': 'F65', 'F39': 'F68', 'F44': 'F89', 'F88': 'F44', 'F93': 'F19', 'F48': 'F9', 'F32': 'F24', 'F9': 'F78', 'F73': 'F83', 'F72': 'F40', 'F71': 'F75', 'F70': 'F36', 'F69': 'F81', 'F68': 'F62', 'F19': 'F58', 'F75': 'F48', 'F10': 'F61', 'F65': 'F27', 'F64': 'F86', 'F11': 'F22', 'F62': 'F73', 'F61': 'F25', 'F60': 'F92', 'F74': 'F3', 'F76': 'F90', 'F58': 'F67', 'F77': 'F41', 'F3': 'F29', 'F4': 'F11', 'F90': 'F60', 'F89': 'F71', 'F5': 'F20', 'F87': 'F21', 'F86': 'F56', 'F85': 'F59', 'F6': 'F35', 'F7': 'F38', 'F82': 'F87', 'F81': 'F5', 'F80': 'F88', 'F79': 'F53', 'F78': 'F4', 'F12': 'F17', 'F56': 'F39', 'F57': 'F45', 'F20': 'F26', 'F36': 'F64', 'F35': 'F23', 'F34': 'F13', 'F33': 'F43', 'F31': 'F79', 'F30': 'F51', 'F29': 'F49', 'F28': 'F7', 'F27': 'F72', 'F26': 'F84', 'F25': 'F32', 'F24': 'F54', 'F23': 'F77', 'F22': 'F6', 'F21': 'F28', 'F37': 'F69', 'F38': 'F50', 'F18': 'F91', 'F49': 'F10', 'F55': 'F70', 'F54': 'F47', 'F53': 'F46', 'F52': 'F31', 'F51': 'F15', 'F50': 'F18', 'F2': 'F52', 'F40': 'F12', 'F14': 'F93', 'F45': 'F76', 'F15': 'F57', 'F43': 'F42', 'F17': 'F37', 'F41': 'F8', 'F1': 'F2'}

{'C2': 'C1', 'C1': 'C2'}

Yes

{'C1': 'No', 'C2': 'Yes'}

GradientBoostingClassifier

C1

Food Ordering Customer Churn Prediction

The prediction probability of C2 is 17.93% and that of C1 is 82.07%. Therefore, the most probable class for the given case is C1. The above classification assertion statements are based on the information supplied to the classifier about the case given. The top features with significant attributions leading to the decision made above are F31, F45, F26, F11, F44, and F27. Conversely, F32, F7, F25, F38, and F2 are among the features deemed irrelevant to the classification decision here since their contributions are almost negligible and much closer to zero. The attribution analysis suggests that not all the relevant features positively contribute to the classifier's arriving at the verdict here. Those with positive attributions that push the classifier towards generating C1 as the label are F31, F45, F26, F11, F9, F30, F29, and F40. Decreasing the likelihood of the correctness of C1 are the negative features such as F44, F34, F27, F16, F39, F6, F19, and F3, which could be blamed for the little uncertainty in the classification output, as indicated by the prediction probability of C2.

{'C2': '17.93%', 'C1': '82.07%'}

{'F31': 'More restaurant choices', 'F45': 'Ease and convenient', 'F26': 'Bad past experience', 'F11': 'Time saving', 'F27': 'Unaffordable', 'F44': 'Educational Qualifications', 'F34': 'Late Delivery', 'F30': 'Occupation', 'F39': 'Influence of rating', 'F6': 'Less Delivery time', 'F16': 'Order placed by mistake', 'F19': 'Delivery person ability', 'F3': 'Order Time', 'F9': 'Unavailability', 'F18': 'More Offers and Discount', 'F4': 'Delay of delivery person picking up food', 'F29': 'Good Taste ', 'F40': 'Wrong order delivered', 'F28': 'Freshness ', 'F23': 'Missing item', 'F32': 'Residence in busy location', 'F25': 'Google Maps Accuracy', 'F7': 'Age', 'F2': 'Good Road Condition', 'F38': 'Low quantity low time', 'F42': 'High Quality of package', 'F24': 'Number of calls', 'F15': 'Politeness', 'F41': 'Temperature', 'F10': 'Maximum wait time', 'F36': 'Long delivery time', 'F8': 'Influence of time', 'F35': 'Delay of delivery person getting assigned', 'F21': 'Family size', 'F1': 'Poor Hygiene', 'F14': 'Health Concern', 'F13': 'Self Cooking', 'F5': 'Good Tracking system', 'F43': 'Good Food quality', 'F46': 'Easy Payment option', 'F37': 'Perference(P2)', 'F20': 'Perference(P1)', 'F22': 'Monthly Income', 'F33': 'Marital Status', 'F17': 'Gender', 'F12': 'Good Quantity'}

{'F12': 'F31', 'F10': 'F45', 'F21': 'F26', 'F11': 'F11', 'F23': 'F27', 'F6': 'F44', 'F19': 'F34', 'F4': 'F30', 'F38': 'F39', 'F39': 'F6', 'F29': 'F16', 'F37': 'F19', 'F31': 'F3', 'F22': 'F9', 'F14': 'F18', 'F26': 'F4', 'F45': 'F29', 'F27': 'F40', 'F43': 'F28', 'F28': 'F23', 'F33': 'F32', 'F34': 'F25', 'F1': 'F7', 'F35': 'F2', 'F36': 'F38', 'F40': 'F42', 'F41': 'F24', 'F42': 'F15', 'F44': 'F41', 'F32': 'F10', 'F24': 'F36', 'F30': 'F8', 'F25': 'F35', 'F7': 'F21', 'F20': 'F1', 'F18': 'F14', 'F17': 'F13', 'F16': 'F5', 'F15': 'F43', 'F13': 'F46', 'F9': 'F37', 'F8': 'F20', 'F5': 'F22', 'F3': 'F33', 'F2': 'F17', 'F46': 'F12'}

{'C1': 'C2', 'C2': 'C1'}

Go Away

{'C2': 'Return', 'C1': 'Go Away'}

DNN

C1

Credit Card Fraud Classification

The classification algorithm classifies the given case as C1 with a confidence level equal to 99.99%, suggesting that there is little chance that the C2 label could be the true label. The classification confidence level can be attributed to the influence and contributions of the features F30, F21, F3, F4, and F7. Positively supporting the model's decision are values of F30, F21, F3, and F4. On the contrary, the values of F7, F12, F11, and F13 are shifting the model towards producing the C2 label, which results in a marginal decrease in the certainty associated with the C1 label. The other positively supported features further improving the odds in favour of C1 include F6, F16, F9, and F27. Overall, it is not farfetched to accept that C1 is the correct label for the case under consideration since the strong positive influences of F30, F21, and F3 far outweigh the influence of any of the other input features. In other words, as mentioned above, there is only a small chance that the true label is not C1 considering the attributions of the top influential input features.

{'C1': '99.99%', 'C2': '0.01%'}

{'F30': 'Z3', 'F21': 'Z6', 'F3': 'Time', 'F7': 'Z13', 'F4': 'Z12', 'F12': 'Z4', 'F13': 'Z10', 'F6': 'Z5', 'F11': 'Z9', 'F24': 'Z14', 'F15': 'Z16', 'F14': 'Z11', 'F16': 'Z17', 'F22': 'Z19', 'F9': 'Z8', 'F23': 'Z28', 'F8': 'Z21', 'F10': 'Z20', 'F5': 'Z1', 'F1': 'Z24', 'F2': 'Z18', 'F27': 'Z2', 'F29': 'Z25', 'F17': 'Amount', 'F25': 'Z26', 'F28': 'Z27', 'F20': 'Z22', 'F26': 'Z15', 'F18': 'Z7', 'F19': 'Z23'}

{'F4': 'F30', 'F7': 'F21', 'F1': 'F3', 'F14': 'F7', 'F13': 'F4', 'F5': 'F12', 'F11': 'F13', 'F6': 'F6', 'F10': 'F11', 'F15': 'F24', 'F17': 'F15', 'F12': 'F14', 'F18': 'F16', 'F20': 'F22', 'F9': 'F9', 'F29': 'F23', 'F22': 'F8', 'F21': 'F10', 'F2': 'F5', 'F25': 'F1', 'F19': 'F2', 'F3': 'F27', 'F26': 'F29', 'F30': 'F17', 'F27': 'F25', 'F28': 'F28', 'F23': 'F20', 'F16': 'F26', 'F8': 'F18', 'F24': 'F19'}

{'C1': 'C1', 'C2': 'C2'}

Not Fraud

{'C1': 'Not Fraud', 'C2': 'Fraud'}

RandomForestClassifier

C2

Employee Attrition

The data is marked as C2 by the classifier based on the input features, with a moderate degree of confidence since the prediction probability of the other label, C1, is only 44.0%. The most influential features driving the classification above are F13, F18, F24, F15, F6, F29, F10, F23, F3, F21, F14, F11, F20, F22, F17, F1, F2, F27, and F16. Strongly reducing the chance of C2 being the true label for the given case are the negative features F18 and F13. Actually, these negative features, along with other features such as F6, F29, and F3, are responsible for the uncertainty in the classification decision here. On the contrary, the input features F24, F15, F10, F23, F21, and F14 positively contribute to the classifier's decision to choose C2 as the label here. Finally, it is important to note that not all the features are shown to be relevant when making the labelling decision regarding the case under consideration, and these irrelevant features include F4, F28, F19, and F26.

{'C1': '44.00%', 'C2': '56.00%'}

{'F13': 'OverTime', 'F18': 'BusinessTravel', 'F24': 'MaritalStatus', 'F15': 'JobInvolvement', 'F6': 'WorkLifeBalance', 'F29': 'Education', 'F10': 'EnvironmentSatisfaction', 'F23': 'Gender', 'F3': 'JobRole', 'F21': 'NumCompaniesWorked', 'F14': 'YearsInCurrentRole', 'F11': 'HourlyRate', 'F20': 'Department', 'F22': 'RelationshipSatisfaction', 'F17': 'PerformanceRating', 'F5': 'YearsWithCurrManager', 'F1': 'Age', 'F2': 'MonthlyRate', 'F27': 'StockOptionLevel', 'F16': 'JobSatisfaction', 'F4': 'DailyRate', 'F30': 'YearsSinceLastPromotion', 'F7': 'YearsAtCompany', 'F26': 'TrainingTimesLastYear', 'F8': 'EducationField', 'F19': 'TotalWorkingYears', 'F25': 'PercentSalaryHike', 'F9': 'MonthlyIncome', 'F12': 'JobLevel', 'F28': 'DistanceFromHome'}

{'F26': 'F13', 'F17': 'F18', 'F25': 'F24', 'F29': 'F15', 'F20': 'F6', 'F27': 'F29', 'F28': 'F10', 'F23': 'F23', 'F24': 'F3', 'F8': 'F21', 'F14': 'F14', 'F4': 'F11', 'F21': 'F20', 'F18': 'F22', 'F19': 'F17', 'F16': 'F5', 'F1': 'F1', 'F7': 'F2', 'F10': 'F27', 'F30': 'F16', 'F2': 'F4', 'F15': 'F30', 'F13': 'F7', 'F12': 'F26', 'F22': 'F8', 'F11': 'F19', 'F9': 'F25', 'F6': 'F9', 'F5': 'F12', 'F3': 'F28'}

{'C1': 'C1', 'C2': 'C2'}

Leave

{'C1': 'Leave', 'C2': 'Leave'}

RandomForestClassifier

C3

Cab Surge Pricing System

The model determined that this case belongs to C3 of the three possible labels, with an 83.0% likelihood. It is important to note, however, that there is about a 14.0% chance that it could be C2 and a 3.0% chance that it is rather C1. The most relevant feature driving this prediction is F5, with a very strong positive attribution, increasing the odds of the label C3. The following attributes have values pushing for a different prediction: F2, F3, F9, and F7, however, their attributions are very low when compared to that from F5. Other features positively contributing to the model's decision for this test case are F6, F12, F8, F11, F4, F1, and F10, with F4, F1, and F10 being the least relevant features considered by the model for the given case.

{'C1': '3.00%', 'C3': '83.00%', 'C2': '14.00%'}

{'F5': 'Type_of_Cab', 'F2': 'Destination_Type', 'F3': 'Trip_Distance', 'F9': 'Cancellation_Last_1Month', 'F6': 'Confidence_Life_Style_Index', 'F12': 'Var3', 'F7': 'Customer_Since_Months', 'F8': 'Life_Style_Index', 'F11': 'Var2', 'F4': 'Gender', 'F1': 'Var1', 'F10': 'Customer_Rating'}

{'F2': 'F5', 'F6': 'F2', 'F1': 'F3', 'F8': 'F9', 'F5': 'F6', 'F11': 'F12', 'F3': 'F7', 'F4': 'F8', 'F10': 'F11', 'F12': 'F4', 'F9': 'F1', 'F7': 'F10'}

{'C3': 'C1', 'C1': 'C3', 'C2': 'C2'}

C2

{'C1': 'Low', 'C3': 'Medium', 'C2': 'High'}

RandomForestClassifier

C1

Annual Income Earnings

The classifier assigned the label C1, given that there is merely a 2.18% chance that C2 is the correct label. Influencing this classification decision are mainly the values of the variables F14, F5, F9, and F2 which are also commonly referred to as positive variables since they increase the response in favour of the predicted label. Other variables supporting the prediction of C1 are F8, F13, F11, and F6. However, unlike F14, F5, F9, and F2, these variables have a moderate impact on the classifier. The variables that decrease the likelihood that C1 is the correct label are F3, F12, F10, and F1 since they have values that swing the classification verdict in the direction of C2.

{'C1': '97.82%', 'C2': '2.18%'}

{'F14': 'Capital Gain', 'F5': 'Marital Status', 'F9': 'Relationship', 'F2': 'Age', 'F8': 'Education-Num', 'F13': 'Hours per week', 'F3': 'Occupation', 'F11': 'Capital Loss', 'F12': 'Sex', 'F6': 'Education', 'F10': 'Race', 'F4': 'fnlwgt', 'F1': 'Country', 'F7': 'Workclass'}

{'F11': 'F14', 'F6': 'F5', 'F8': 'F9', 'F1': 'F2', 'F5': 'F8', 'F13': 'F13', 'F7': 'F3', 'F12': 'F11', 'F10': 'F12', 'F4': 'F6', 'F9': 'F10', 'F3': 'F4', 'F14': 'F1', 'F2': 'F7'}

{'C1': 'C1', 'C2': 'C2'}

Under 50K

{'C1': 'Under 50K', 'C2': 'Above 50K'}

SVMClassifier_poly

C1

Employee Attrition

Because the chance that C2 is the right label is around 42.17 percent, the example under review is labelled as C1 with a moderate degree of confidence. F30, F18, F2, F10, F12, and F3 have the most influence on the above forecast, whereas F24, F14, F13, F7, F8, F15, and F9 have small contributions. F19, F21, F1, F16, F4, F22, and F5 all have a relatively modest impact. However, the classifier does not take into account all of the attributes while making a judgement in a specific case and the attributes F29, F23, F6, and F26 are all irrelevant features. F30, F18, F2, F12, F14, F8, and F22 are the positive features pushing the prediction in support of the forecasted label. We can see from the attributions map that the bulk of the influential features exhibit negative attributions that reduce the likelihood that C1 is the correct label, justifying the uncertainty associated with the classifier's prediction choice.

{'C1': '57.83%', 'C2': '42.17%'}

{'F30': 'OverTime', 'F18': 'NumCompaniesWorked', 'F2': 'RelationshipSatisfaction', 'F10': 'MaritalStatus', 'F12': 'YearsSinceLastPromotion', 'F3': 'Department', 'F24': 'Age', 'F14': 'Education', 'F13': 'EducationField', 'F7': 'BusinessTravel', 'F8': 'JobLevel', 'F15': 'JobInvolvement', 'F9': 'WorkLifeBalance', 'F19': 'MonthlyRate', 'F21': 'YearsAtCompany', 'F1': 'Gender', 'F16': 'PerformanceRating', 'F4': 'JobRole', 'F22': 'TrainingTimesLastYear', 'F5': 'EnvironmentSatisfaction', 'F29': 'YearsWithCurrManager', 'F23': 'DailyRate', 'F6': 'YearsInCurrentRole', 'F26': 'TotalWorkingYears', 'F28': 'StockOptionLevel', 'F17': 'PercentSalaryHike', 'F11': 'MonthlyIncome', 'F20': 'HourlyRate', 'F25': 'DistanceFromHome', 'F27': 'JobSatisfaction'}

{'F26': 'F30', 'F8': 'F18', 'F18': 'F2', 'F25': 'F10', 'F15': 'F12', 'F21': 'F3', 'F1': 'F24', 'F27': 'F14', 'F22': 'F13', 'F17': 'F7', 'F5': 'F8', 'F29': 'F15', 'F20': 'F9', 'F7': 'F19', 'F13': 'F21', 'F23': 'F1', 'F19': 'F16', 'F24': 'F4', 'F12': 'F22', 'F28': 'F5', 'F16': 'F29', 'F2': 'F23', 'F14': 'F6', 'F11': 'F26', 'F10': 'F28', 'F9': 'F17', 'F6': 'F11', 'F4': 'F20', 'F3': 'F25', 'F30': 'F27'}

{'C1': 'C1', 'C2': 'C2'}

Leave

{'C1': 'Leave', 'C2': 'Leave'}

BernoulliNB

C1

Credit Card Fraud Classification

All features are shown to have a positive impact on the classification to class C1 or to have no impact at all. F14, F11, F22, and F27 are the four features with the most impact. Some of the remaining features, in order of feature importance, are F4, F2, F18, F9, F30, F10, F5, F15, F12, F25, F16, and F8. F14 and F11 both have the highest positive impact on the final classification, pushing the classification towards class C1. All of F22, F27, F4, and F2 influence the model's classification to C1. In terms of the features which have a positive impact on the classification, features F18, F9, F30, and F10 are all ranked to have a medium degree of influence on the final classification. F18 and F9 both have a similar importance attribution, which is higher than that of F30 and F10. All the other features not listed above are irrelevant to the decision above and among them are F1, F3, and F24.

{'C2': '5.16%', 'C1': '94.84%'}

{'F14': 'Z14', 'F11': 'Z1', 'F22': 'Z17', 'F27': 'Amount', 'F4': 'Z19', 'F2': 'Z5', 'F18': 'Z3', 'F9': 'Z8', 'F30': 'Z18', 'F10': 'Z10', 'F5': 'Z26', 'F15': 'Z25', 'F12': 'Z22', 'F25': 'Z4', 'F16': 'Z7', 'F8': 'Z13', 'F19': 'Z23', 'F7': 'Z9', 'F23': 'Z21', 'F20': 'Z2', 'F1': 'Z28', 'F24': 'Z24', 'F3': 'Z27', 'F29': 'Time', 'F17': 'Z20', 'F26': 'Z16', 'F13': 'Z12', 'F6': 'Z11', 'F28': 'Z6', 'F21': 'Z15'}

{'F15': 'F14', 'F2': 'F11', 'F18': 'F22', 'F30': 'F27', 'F20': 'F4', 'F6': 'F2', 'F4': 'F18', 'F9': 'F9', 'F19': 'F30', 'F11': 'F10', 'F27': 'F5', 'F26': 'F15', 'F23': 'F12', 'F5': 'F25', 'F8': 'F16', 'F14': 'F8', 'F24': 'F19', 'F10': 'F7', 'F22': 'F23', 'F3': 'F20', 'F29': 'F1', 'F25': 'F24', 'F28': 'F3', 'F1': 'F29', 'F21': 'F17', 'F17': 'F26', 'F13': 'F13', 'F12': 'F6', 'F7': 'F28', 'F16': 'F21'}

{'C2': 'C2', 'C1': 'C1'}

Fraud

{'C2': 'Not Fraud', 'C1': 'Fraud'}

LogisticRegression

C2

Used Cars Price-Range Prediction

Label C1 has a lower probability than label C2, so C2 is the most likely option in this case. C2 has a probability of approximately 96.25 percent, which can be attributed to variables such as F9, F7, F1, and F3. According to the attributions assessment, the least relevant variables are F2, F4, and F10. Inspection of the direction of influence of the features showed that F5 and F8 present negative contributions that push the model somewhat away from producing C2 because they support the label C1. Considering that the combined impact of the negative variables is quite minimal in comparison to the combined impact of the positive variables such as F9, F7, F1, F6, and F3, it is not surprising that the model is very certain that C1 is not the accurate label for the given case.

{'C2': '96.25%', 'C1': '3.75%'}

{'F7': 'Fuel_Type', 'F9': 'Power', 'F1': 'Engine', 'F3': 'Seats', 'F5': 'car_age', 'F8': 'Owner_Type', 'F6': 'Name', 'F2': 'Mileage', 'F4': 'Kilometers_Driven', 'F10': 'Transmission'}

{'F7': 'F7', 'F4': 'F9', 'F3': 'F1', 'F10': 'F3', 'F5': 'F5', 'F9': 'F8', 'F6': 'F6', 'F2': 'F2', 'F1': 'F4', 'F8': 'F10'}

{'C1': 'C2', 'C2': 'C1'}

Low

{'C2': 'Low', 'C1': 'High'}

GradientBoostingClassifier

C1

Paris House Classification

According to the prediction made here, the most likely label for the given case is C1, with a prediction probability of 97.02%, indicating that the prediction probability of C2 is only 2.98%. The classification above is mainly due to the influence of F1, F3, and F12. The next set of features with moderate contributions includes F16, F4, and F7. However, those with little consideration from the classifier are F6, F13, F2, and F17. In consideration of the fact that all the top four features have a strong positive contribution, it is foreseeable why the classifier is relatively confident that the correct label for this case is C1. Additionally, the negative attributes with moderate to low impact are F4, F5, and F11.

{'C2': '2.98%', 'C1': '97.02%'}

{'F1': 'isNewBuilt', 'F3': 'hasYard', 'F12': 'hasPool', 'F16': 'hasStormProtector', 'F4': 'made', 'F7': 'squareMeters', 'F5': 'floors', 'F11': 'cityCode', 'F10': 'hasGuestRoom', 'F8': 'basement', 'F9': 'numPrevOwners', 'F14': 'price', 'F15': 'numberOfRooms', 'F6': 'garage', 'F13': 'cityPartRange', 'F2': 'hasStorageRoom', 'F17': 'attic'}

{'F3': 'F1', 'F1': 'F3', 'F2': 'F12', 'F4': 'F16', 'F12': 'F4', 'F6': 'F7', 'F8': 'F5', 'F9': 'F11', 'F16': 'F10', 'F13': 'F8', 'F11': 'F9', 'F17': 'F14', 'F7': 'F15', 'F15': 'F6', 'F10': 'F13', 'F5': 'F2', 'F14': 'F17'}

{'C1': 'C2', 'C2': 'C1'}

Luxury

{'C2': 'Basic', 'C1': 'Luxury'}

RandomForestClassifier

C2

Flight Price-Range Classification

The classification conclusion is as follows: C2 is the most likely label for this case and the classifier is certain that neither C3 nor C1 are the right labels since their likelihoods are equal to zero. The driving factors for the above classification are F4, F9, and F1, all of which have a substantial positive impact, causing the classifier to select C2. F8, F3, F2, and F10 are also positive features. The assigned label is not supported by all of the input features since the negative features F5, F7, and F12 support the decision that the most likely class for this instance could be any one of the other labels, C1 and C3. Nevertheless, given the confidence level in the aforementioned classification, it is reasonable to assume that the classifier paid little attention to the negative features, resulting in the selection of class C2.

{'C2': '100.00%', 'C3': '0.00%', 'C1': '0.00%'}

{'F4': 'Airline', 'F9': 'Duration_hours', 'F1': 'Total_Stops', 'F8': 'Journey_month', 'F5': 'Source', 'F3': 'Destination', 'F2': 'Arrival_hour', 'F7': 'Journey_day', 'F12': 'Dep_minute', 'F10': 'Arrival_minute', 'F11': 'Duration_mins', 'F6': 'Dep_hour'}

{'F9': 'F4', 'F7': 'F9', 'F12': 'F1', 'F2': 'F8', 'F10': 'F5', 'F11': 'F3', 'F5': 'F2', 'F1': 'F7', 'F4': 'F12', 'F6': 'F10', 'F8': 'F11', 'F3': 'F6'}

{'C1': 'C2', 'C3': 'C3', 'C2': 'C1'}

Low

{'C2': 'Low', 'C3': 'Moderate', 'C1': 'High'}

KNeighborsClassifier

C1

Water Quality Classification

The classifier states that there is a 50.0% chance that the true label of this test observation is C1. This indicates that the classifier is less certain in its prediction decision regarding the case under consideration. The label assigned is mainly due to the values of the features F5, F7, F2, F4, F1, and F3. The top features, F5 and F7, have very strong positive contributions pushing the prediction higher towards the most probable label. Among the remaining features stated above, F2, F4, F1, and F3, only F3 demonstrates some level of contradiction, forcing the labelling decision in a different direction. Finally, the features with marginal impact on the prediction made here are F6, F9, and F8. While F6 and F8 positively influence the decision made, F9 suggests that the label assigned by the classifier might not be the true label.

{'C1': '50.00%', 'C2': '50.00%'}

{'F5': 'Hardness', 'F7': 'Sulfate', 'F2': 'Organic_carbon', 'F4': 'Solids', 'F1': 'Conductivity', 'F3': 'Trihalomethanes', 'F6': 'ph', 'F9': 'Turbidity', 'F8': 'Chloramines'}

{'F2': 'F5', 'F5': 'F7', 'F7': 'F2', 'F3': 'F4', 'F6': 'F1', 'F8': 'F3', 'F1': 'F6', 'F9': 'F9', 'F4': 'F8'}

{'C2': 'C1', 'C1': 'C2'}

Not Portable

{'C1': 'Not Portable', 'C2': 'Portable'}

GradientBoostingClassifier

C2

Printer Sales

According to the attribution analysis, the each input variables contributes differently to the decision. For the case under consideration, there are variables that have negative influence on the decision here, but it also has numerous quantifiable variables that are positive. Per the model, C2 is 91.95% certain to be the correct label and C1 has a predicted probability of only 8.05%. The most essential input variables are F17, F2, F11, and F21, which allow the model to effectively compute the likelihoods across the classes, C2 and C1. F25 and F6 have nearly comparable positive effects, but F22 and F7 have a negative influence, altering the output decision in favour of a different label. The cumulative positive contribution of F25, F17, F21, F3, F10, and F6 was greater than that of F2, F11, F7, and F22, hence the positive variables succeed at improving the predictability odds in favour of the C2 class. Furthermore per the variable attributions, the contributions of F4, F18, F9, and F1 has very little to do with the classification decision since their attributions are negligible and closer to zero than all the above-mentioned variables.

{'C1': '8.05%', 'C2': '91.95%'}

{'F17': 'X24', 'F2': 'X8', 'F21': 'X1', 'F11': 'X21', 'F7': 'X4', 'F6': 'X6', 'F25': 'X3', 'F22': 'X22', 'F3': 'X7', 'F10': 'X15', 'F8': 'X20', 'F15': 'X11', 'F26': 'X10', 'F23': 'X19', 'F13': 'X5', 'F19': 'X16', 'F12': 'X23', 'F20': 'X9', 'F14': 'X17', 'F24': 'X18', 'F16': 'X25', 'F5': 'X14', 'F1': 'X2', 'F9': 'X13', 'F18': 'X12', 'F4': 'X26'}

{'F24': 'F17', 'F8': 'F2', 'F1': 'F21', 'F21': 'F11', 'F4': 'F7', 'F6': 'F6', 'F3': 'F25', 'F22': 'F22', 'F7': 'F3', 'F15': 'F10', 'F20': 'F8', 'F11': 'F15', 'F10': 'F26', 'F19': 'F23', 'F5': 'F13', 'F16': 'F19', 'F23': 'F12', 'F9': 'F20', 'F17': 'F14', 'F18': 'F24', 'F25': 'F16', 'F14': 'F5', 'F2': 'F1', 'F13': 'F9', 'F12': 'F18', 'F26': 'F4'}

{'C1': 'C1', 'C2': 'C2'}

More

{'C1': 'Less', 'C2': 'More'}

LogisticRegression

C2

Annual Income Earnings

Deciding the most probable label for the given case on the basis of the values of the input variables, the classification algorithm's output decision is that: the probability of C2 being the correct label is 79.78%, the probability of C1 is 20.22%. Therefore, the most likely label is identified as C2 and the attribution analysis shows that all the variables contributed to some extent to the final decision by the algorithm with respect to the given case. The most influential variables are F9, F7, F6, and F2, but F1, F12, and F14 are the least influential ones. The analysis also indicates that F9, F2, F1, and F14 are responsible for the marginal doubt in the classification decision here hence they are commonly referred to as negative variables since their contributions only tend to shift the verdict in a different direction than the assigned label. Finally, the variables such as F7, F6, F13, F8, F10, and F5 are the positive variables that increase the algorithm's response in favour of outputting the C2 label.

{'C1': '20.22%', 'C2': '79.78%'}

{'F9': 'Capital Gain', 'F7': 'Marital Status', 'F6': 'Education', 'F2': 'Capital Loss', 'F13': 'Hours per week', 'F8': 'Sex', 'F10': 'Country', 'F5': 'Education-Num', 'F11': 'Occupation', 'F3': 'Race', 'F4': 'Age', 'F1': 'Workclass', 'F12': 'fnlwgt', 'F14': 'Relationship'}

{'F11': 'F9', 'F6': 'F7', 'F4': 'F6', 'F12': 'F2', 'F13': 'F13', 'F10': 'F8', 'F14': 'F10', 'F5': 'F5', 'F7': 'F11', 'F9': 'F3', 'F1': 'F4', 'F2': 'F1', 'F3': 'F12', 'F8': 'F14'}

{'C1': 'C1', 'C2': 'C2'}

Above 50K

{'C1': 'Under 50K', 'C2': 'Above 50K'}

LogisticRegression

C2

E-Commerce Shipping

The reliability of the classification verdict for this case is 71.57%, implying there is a 28.43% chance that the correct label could be C1. F10 has a significant negative impact on classification output since its contribution contradicts the labelling of the case as C2, hence favours labelling the case as C1. The values F6, F3, F5, F4, F9, F1, and F8 have a positive effect on the results, but still contributes less than the effect of F10. The analysis shows that F10 has an overwhelming negative impact or influence on the predictive decisions of the model here. F3, F5, F4, and F9 have a positive effect on model predictions. Due to the power of the F10 function, all other functions have little effect on the results. In summary, the uncertainty of the predictions can be explained by the control on the model by feature F10, which drags the classification decision favourably towards C1.

{'C2': '71.57%', 'C1': '28.43%'}

{'F10': 'Discount_offered', 'F6': 'Weight_in_gms', 'F3': 'Prior_purchases', 'F5': 'Product_importance', 'F4': 'Cost_of_the_Product', 'F9': 'Gender', 'F1': 'Customer_rating', 'F8': 'Warehouse_block', 'F2': 'Customer_care_calls', 'F7': 'Mode_of_Shipment'}

{'F2': 'F10', 'F3': 'F6', 'F8': 'F3', 'F9': 'F5', 'F1': 'F4', 'F10': 'F9', 'F7': 'F1', 'F4': 'F8', 'F6': 'F2', 'F5': 'F7'}

{'C2': 'C2', 'C1': 'C1'}

On-time

{'C2': 'On-time', 'C1': 'Late'}

DecisionTreeClassifier

C1

Vehicle Insurance Claims

This model predicted class label C1 with about 93.32% certainty, while there was about a 6.68% chance of the correct class being identified as a different label. Seven features, F29, F15, F5, F10, F19, F23, and F7, have higher impacts on the model prediction decision above. But the feature F29 has the largest positive impact on the result and on the contrary, F5, F19, F23, and F7 show the potential to shift the narrative to a different label since their contributions reduce the likelihood of the predicted label for this case. In addition, features F8, F33, and F27 have moderate impacts on the model's prediction but each of them is increasing the responses, and finally, the features shown have negligible influence include F9, and F1.

{'C1': '93.32%', 'C2': '6.68%'}

{'F29': 'incident_severity', 'F15': 'incident_city', 'F5': 'injury_claim', 'F10': 'insured_occupation', 'F19': 'insured_zip', 'F23': 'authorities_contacted', 'F7': 'auto_year', 'F8': 'police_report_available', 'F33': 'bodily_injuries', 'F27': 'insured_hobbies', 'F30': 'insured_sex', 'F14': 'auto_make', 'F26': 'property_damage', 'F17': 'witnesses', 'F4': 'insured_relationship', 'F24': 'age', 'F16': 'vehicle_claim', 'F11': 'months_as_customer', 'F6': 'property_claim', 'F13': 'incident_type', 'F20': 'capital-gains', 'F21': 'policy_deductable', 'F28': 'policy_annual_premium', 'F31': 'incident_state', 'F18': 'umbrella_limit', 'F32': 'total_claim_amount', 'F22': 'collision_type', 'F2': 'incident_hour_of_the_day', 'F3': 'insured_education_level', 'F9': 'number_of_vehicles_involved', 'F1': 'policy_csl', 'F25': 'policy_state', 'F12': 'capital-loss'}

{'F27': 'F29', 'F30': 'F15', 'F14': 'F5', 'F22': 'F10', 'F6': 'F19', 'F28': 'F23', 'F17': 'F7', 'F32': 'F8', 'F11': 'F33', 'F23': 'F27', 'F20': 'F30', 'F33': 'F14', 'F31': 'F26', 'F12': 'F17', 'F24': 'F4', 'F2': 'F24', 'F16': 'F16', 'F1': 'F11', 'F15': 'F6', 'F25': 'F13', 'F7': 'F20', 'F3': 'F21', 'F4': 'F28', 'F29': 'F31', 'F5': 'F18', 'F13': 'F32', 'F26': 'F22', 'F9': 'F2', 'F21': 'F3', 'F10': 'F9', 'F19': 'F1', 'F18': 'F25', 'F8': 'F12'}

{'C1': 'C1', 'C2': 'C2'}

Not Fraud

{'C1': 'Not Fraud', 'C2': 'Fraud'}

KNeighborsClassifier

C1

Printer Sales

Considering the prediction likelihoods, this case is labelled as C1 by the model, that is, the model states that there is about an 83.33% chance that the case is under C1 and about a 16.67% chance that it is not. The most relevant features influencing the decision made here are: F24, F17, F5, and F18. Among the feature set mentioned above, F24 and F17 offer a very strong positive contribution to the prediction of C1. Conversely, F18 suggests the alternative label C2 could be the true label for this case, but this attribution is weak when compared to F17 and F24. Other features that are moderately pushing for this classification decision include F23, F26, F7, and F21. However, the values of F1, F19, F3, and F20 advocate for the assignment of a different label. Finally, the features F16, F2, F9, and F15 have very little impact on the model's prediction for this case.

{'C1': '83.33%', 'C2': '16.67%'}

{'F24': 'X24', 'F17': 'X1', 'F18': 'X4', 'F5': 'X10', 'F23': 'X2', 'F26': 'X8', 'F1': 'X17', 'F7': 'X7', 'F21': 'X21', 'F19': 'X18', 'F3': 'X6', 'F20': 'X11', 'F22': 'X22', 'F14': 'X25', 'F25': 'X5', 'F13': 'X19', 'F8': 'X15', 'F4': 'X23', 'F10': 'X16', 'F11': 'X3', 'F16': 'X14', 'F2': 'X20', 'F9': 'X13', 'F15': 'X12', 'F12': 'X9', 'F6': 'X26'}

{'F24': 'F24', 'F1': 'F17', 'F4': 'F18', 'F10': 'F5', 'F2': 'F23', 'F8': 'F26', 'F17': 'F1', 'F7': 'F7', 'F21': 'F21', 'F18': 'F19', 'F6': 'F3', 'F11': 'F20', 'F22': 'F22', 'F25': 'F14', 'F5': 'F25', 'F19': 'F13', 'F15': 'F8', 'F23': 'F4', 'F16': 'F10', 'F3': 'F11', 'F14': 'F16', 'F20': 'F2', 'F13': 'F9', 'F12': 'F15', 'F9': 'F12', 'F26': 'F6'}

{'C2': 'C1', 'C1': 'C2'}

Less

{'C1': 'Less', 'C2': 'More'}

SVC

C1

Advertisement Prediction

For the given instance, the model generated the label C1 with a very high predicted probability equal to 99.66% which implies that the model is very confident that C2 is not the correct label. Ranking the contributions of the features to the prediction above, from the most relevant to the least relevant, is as follows: F6, F7, F5, F2, F4, F1, and F3. Among the seven features, only F4 and F1 have negative contributions, pushing the prediction towards the C2 label. However, given that these features have very low contributions, their impact on the model's decision is close to negligible when compared to the contributions of the positive features F6, F7, and F5.

{'C2': '0.34%', 'C1': '99.66%'}

{'F6': 'Daily Time Spent on Site', 'F7': 'Daily Internet Usage', 'F5': 'Age', 'F2': 'Gender', 'F4': 'ad_day', 'F1': 'ad_month', 'F3': 'Area Income'}

{'F1': 'F6', 'F4': 'F7', 'F2': 'F5', 'F5': 'F2', 'F7': 'F4', 'F6': 'F1', 'F3': 'F3'}

{'C2': 'C2', 'C1': 'C1'}

Watch

{'C2': 'Skip', 'C1': 'Watch'}

LogisticRegression

C2

Student Job Placement

The final prediction given by the model was C2 with almost 100% certainty, showing the model is confident about its decision. F8 had significantly more influence on the prediction than any other feature with F7 and F12 having the next highest attribution values. All the top features, F8, F7, and F12, encouraged the model to output class C2. F4, F11, and F3 are the features that had the least positive impact on the final classification. The features F5, F6, F9, and F1 have moderate impacts, pushing the model slightly away from a C2 classification.

{'C2': '98.47%', 'C1': '1.53%'}

{'F8': 'ssc_p', 'F7': 'hsc_p', 'F12': 'degree_p', 'F10': 'gender', 'F5': 'degree_t', 'F6': 'workex', 'F2': 'specialisation', 'F9': 'etest_p', 'F4': 'hsc_s', 'F1': 'hsc_b', 'F11': 'ssc_b', 'F3': 'mba_p'}

{'F1': 'F8', 'F2': 'F7', 'F3': 'F12', 'F6': 'F10', 'F10': 'F5', 'F11': 'F6', 'F12': 'F2', 'F4': 'F9', 'F9': 'F4', 'F8': 'F1', 'F7': 'F11', 'F5': 'F3'}

{'C1': 'C2', 'C2': 'C1'}

Not Placed

{'C2': 'Not Placed', 'C1': 'Placed'}

GradientBoostingClassifier

C1

Health Care Services Satisfaction Prediction

Based on the information provided to the classifier, the true label for the given case is likely C1, with a confidence level of 76.26%. Each input variable has a different degree of influence on the classifier's final labelling decision with respect to the case under consideration. Whilst F11, F13, and F7 have lower contributions to the classifier's decision, F16, F2, and F6 are identified as the major contributors resulting in the assignment and classification probabilities across the two classes. There is a 23.74% chance that perhaps C2 is the true label and the features responsible for this are the negative features, F6, F14, F1, F15, F5, F11, and F13. Driving the classifier's decision in favour of C1 are the positive features such as F16, F2, F8, F12, F9, F3, and F10.

{'C2': '23.74%', 'C1': '76.26%'}

{'F16': 'Exact diagnosis', 'F2': 'avaliablity of drugs', 'F6': 'lab services', 'F8': 'friendly health care workers', 'F12': 'Communication with dr', 'F14': 'Time waiting', 'F9': 'Specialists avaliable', 'F3': 'Modern equipment', 'F10': 'waiting rooms', 'F1': 'Check up appointment', 'F4': 'Hygiene and cleaning', 'F15': 'Admin procedures', 'F5': 'Time of appointment', 'F11': 'hospital rooms quality', 'F13': 'parking, playing rooms, caffes', 'F7': 'Quality\\/experience dr.'}

{'F9': 'F16', 'F13': 'F2', 'F12': 'F6', 'F11': 'F8', 'F8': 'F12', 'F2': 'F14', 'F7': 'F9', 'F10': 'F3', 'F14': 'F10', 'F1': 'F1', 'F4': 'F4', 'F3': 'F15', 'F5': 'F5', 'F15': 'F11', 'F16': 'F13', 'F6': 'F7'}

{'C2': 'C2', 'C1': 'C1'}

Satisfied

{'C2': 'Dissatisfied', 'C1': 'Satisfied'}

SGDClassifier

C3

Flight Price-Range Classification

According to the model, C1 is the least probable class, while the most probable class for the given case is identified as C3. The top two variables with the greatest control over the model in terms of this case's label assignment are F8 and F7 but on the contrary, the rest of the variables have moderate-to-lower influence. The contribution of F7 is negative, reducing the chances of selecting the label C3. F8, F3, and F10 drive the model to classify the given case as C3. Furthermore, both F4 and F2 have values that increase the predicted probability of C3, but F6 and F11 decrease the model's response in favour of any of the remaining classes. When choosing a label in this instance, the model pays little attention to the respective values of F9, F5, F12, and F1 hence they are the least ranked features.

{'C3': '86.54%', 'C2': '13.46%', 'C1': '0.00%'}

{'F8': 'Airline', 'F7': 'Total_Stops', 'F3': 'Source', 'F10': 'Journey_month', 'F6': 'Arrival_minute', 'F4': 'Journey_day', 'F2': 'Duration_hours', 'F11': 'Dep_hour', 'F9': 'Destination', 'F5': 'Arrival_hour', 'F12': 'Dep_minute', 'F1': 'Duration_mins'}

{'F9': 'F8', 'F12': 'F7', 'F10': 'F3', 'F2': 'F10', 'F6': 'F6', 'F1': 'F4', 'F7': 'F2', 'F3': 'F11', 'F11': 'F9', 'F5': 'F5', 'F4': 'F12', 'F8': 'F1'}

{'C2': 'C3', 'C1': 'C2', 'C3': 'C1'}

Low

{'C3': 'Low', 'C2': 'Moderate', 'C1': 'High'}

SVC

C2

Australian Credit Approval

Judging by the prediction probabilities, the most probable or likely class assigned by the classifier is C2, with the associated confidence level of 90.97%. The features with the most influence on the prediction above include F10, F6, and F2, while the least important features are F13, F8, and F4. Beside some of the features are shown to negatively contribute to the prediction made here and these negative features, F11, F6, F12, F1, and F13, reduce the classifier's response to generating label C2, consequently pushing the verdict towards C1. The joint impact of the negatives is smaller compared to that of positive features such as F10, F2, F9, and F3, hence the greater drive on the classifier to assign C2 as the correct label.

{'C1': '9.03%', 'C2': '90.97%'}

{'F10': 'A8', 'F6': 'A9', 'F2': 'A12', 'F11': 'A10', 'F9': 'A4', 'F12': 'A14', 'F1': 'A11', 'F3': 'A13', 'F7': 'A1', 'F5': 'A6', 'F14': 'A3', 'F13': 'A5', 'F8': 'A2', 'F4': 'A7'}

{'F8': 'F10', 'F9': 'F6', 'F12': 'F2', 'F10': 'F11', 'F4': 'F9', 'F14': 'F12', 'F11': 'F1', 'F13': 'F3', 'F1': 'F7', 'F6': 'F5', 'F3': 'F14', 'F5': 'F13', 'F2': 'F8', 'F7': 'F4'}

{'C2': 'C1', 'C1': 'C2'}

Class 2

{'C1': 'Class 1', 'C2': 'Class 2'}

KNeighborsClassifier

C2

Credit Risk Classification

According to the model employed, the label for the case is more likely to be C2. This assessment decision is mainly based on the inpacts of features such as F4, F2, F3, F10, and F1. Among these top features, F4, F2, and F3 have positive contributions to the prediction above, while F1 and F10 are identified as negative features which decreases the likelihood associated with class C2 for this case. Furthermore, the values of F7, F6, F11, and F9 also indicate that the other label, C1, may be the correct label but luckily, the influence of the above-mentioned negative features can be classified as only moderate when compared to F4, F2, and F3. In conclusion, with such a strong positive influence from F4, F2, F3, and F5, it is safe to say that the model is very accurate in its classification judgments, with 100.0% certainty.

{'C2': '100.00%', 'C1': '0.00%'}

{'F4': 'fea_4', 'F2': 'fea_8', 'F3': 'fea_2', 'F1': 'fea_9', 'F10': 'fea_6', 'F11': 'fea_10', 'F9': 'fea_1', 'F5': 'fea_7', 'F6': 'fea_11', 'F7': 'fea_3', 'F8': 'fea_5'}

{'F4': 'F4', 'F8': 'F2', 'F2': 'F3', 'F9': 'F1', 'F6': 'F10', 'F10': 'F11', 'F1': 'F9', 'F7': 'F5', 'F11': 'F6', 'F3': 'F7', 'F5': 'F8'}

{'C1': 'C2', 'C2': 'C1'}

Low

{'C2': 'Low', 'C1': 'High'}

BernoulliNB

C1

Personal Loan Modelling

From the prediction likelihood of each class label, the most probable label for the given case based on the values of its features is C1. The likelihood of C2 is negligible, hence we can conclude that the classifier is very confident that C1 is the correct label. Analysing the attributions of the input features showed that the most relevant feature with a strong influence on the classifier's decision here is F5. However, the classifier likely disregards the values of the irrelevant features, F2 and F8, when arriving at the classification above. The confidence level of the classifier employed to make the classification decision above is higher, mainly because the majority of the influential features have positive contributions. Positive features such as F5, F7, and F3 increase the classifier's response higher in favour of C1. F9 and F1 are the main negative features, but compared to F5, their influence on the above classification is very small.

{'C1': '99.99%', 'C2': '0.01%'}

{'F5': 'CD Account', 'F9': 'Income', 'F3': 'CCAvg', 'F7': 'Securities Account', 'F1': 'Education', 'F4': 'Family', 'F6': 'Mortgage', 'F2': 'Age', 'F8': 'Extra_service'}

{'F8': 'F5', 'F2': 'F9', 'F4': 'F3', 'F7': 'F7', 'F5': 'F1', 'F3': 'F4', 'F6': 'F6', 'F1': 'F2', 'F9': 'F8'}

{'C2': 'C1', 'C1': 'C2'}

Reject

{'C1': 'Reject', 'C2': 'Accept'}

LogisticRegression

C1

Bike Sharing Demand

The correct label for the given data instance, according to the machine learning algorithm, is C1 and this is mainly because the probability that C2 is the right label is only about 3.08%. From the analysis, the ranking of the input features based on their respective degree of influence is F11, F4, F9, F2, F8, F5, F12, F6, F3, F1, F7, and F10. This implies the most relevant features are F11, and F4 whereas F7 and F10 are the least relevant ones. Given that F3, F1, and F7 are the features that have a negative impact on the algorithm's selection in this case, it's no wonder that it's quite confident in the chosen class. The arguement towards labelling the case as C1 is also supported by the fact that the joint negative contributions of F3, F1, and F7 is very small when compared to that of the top positive features F11, F4, F9, F2, and F8.

{'C2': '3.08%', 'C1': '96.92%'}

{'F11': 'Functioning Day', 'F4': 'Rainfall(mm)', 'F9': 'Snowfall (cm)', 'F2': 'Solar Radiation (MJ\\/m2)', 'F8': 'Temperature', 'F5': 'Holiday', 'F12': 'Humidity(%)', 'F6': 'Seasons', 'F3': 'Hour', 'F1': 'Visibility (10m)', 'F7': 'Dew point temperature', 'F10': 'Wind speed (m\\/s)'}

{'F12': 'F11', 'F8': 'F4', 'F9': 'F9', 'F7': 'F2', 'F2': 'F8', 'F11': 'F5', 'F3': 'F12', 'F10': 'F6', 'F1': 'F3', 'F5': 'F1', 'F6': 'F7', 'F4': 'F10'}

{'C1': 'C2', 'C2': 'C1'}

More than 500

{'C2': 'Less than 500', 'C1': 'More than 500'}

MLPClassifier

C2

Hotel Satisfaction

Based on the values of the input variables, the prediction model labels the case given as C2 with very high certainty. Specifically, there is only about a 5.59% possibility that C1 is the correct label according to the model. The most influential factors leading to the above prediction decision are the values of F1, F9, and F7 whereas F11, F6, and F3 are deemed less relevant by the model. In between the two ends (most influential and least influential) are the features such as F4, F2, and F8 with moderate contributions. According to the attribution investigation performed, F1, F2, F8, F15, F5, and F3 have positive contributions, increasing the model's response to favour labelling the case as "C2". Conversely, features such as F9, F7, F10, and F4 provide negative contributions, resulting in a small shift toward selecting C1 as the correct class. In conclusion, given that the prediction likelihood of C1 is only 5.59%, it is obvious that the positive features outweigh the negative ones in terms of the considerations they receive from the model, hence the model's decision to assign the C2 label.

{'C2': '94.41%', 'C1': '5.59%'}

{'F1': 'Hotel wifi service', 'F9': 'Type of Travel', 'F7': 'Other service', 'F4': 'Stay comfort', 'F2': 'Type Of Booking', 'F8': 'Ease of Online booking', 'F10': 'Checkin\\/Checkout service', 'F15': 'Age', 'F5': 'Cleanliness', 'F12': 'Food and drink', 'F13': 'Hotel location', 'F14': 'Departure\\/Arrival convenience', 'F11': 'Gender', 'F6': 'purpose_of_travel', 'F3': 'Common Room entertainment'}

{'F6': 'F1', 'F3': 'F9', 'F14': 'F7', 'F11': 'F4', 'F4': 'F2', 'F8': 'F8', 'F13': 'F10', 'F5': 'F15', 'F15': 'F5', 'F10': 'F12', 'F9': 'F13', 'F7': 'F14', 'F1': 'F11', 'F2': 'F6', 'F12': 'F3'}

{'C2': 'C2', 'C1': 'C1'}

dissatisfied

{'C2': 'dissatisfied', 'C1': 'satisfied'}

LogisticRegression

C2

Airline Passenger Satisfaction

The data under consideration is labelled as C2 since it is the most probable label, with a prediction likelihood equal to 99.97% therefore classifier employed here is very confident that C1 is not the right label. The top features with the greatest influence on the classifier in terms of the above classification are F15, F6, F13, and F19. Conversely, the values of F14 and F7 have inconsiderable or insignificant influence on the decision made by the classifier. The input features with moderate to low influence but higher than F14 and F7 on the classifier include F22, F8, F16, and F11. The analysis also shows that the majority of the input features have positive attributions, explaining the level of confidence of the classifier as demonstrated by the prediction probabilities across the classes. The positive features increasing the odds of being labelled C2 include F15, F6, F13, F22, F8, F16, and F9. The marginal doubt in the prediction made here could be attributed to the influence of negative features such as F19, F11, F2, and F3. The negative features support classifying the given data as C1, but since their collective influence is smaller compared to that of the positives, the classifier is shifted more towards labelling the data as C2.

{'C1': '0.03%', 'C2': '99.97%'}

{'F6': 'Inflight wifi service', 'F15': 'Type of Travel', 'F13': 'Customer Type', 'F19': 'Online boarding', 'F22': 'Inflight service', 'F8': 'Baggage handling', 'F16': 'On-board service', 'F11': 'Departure\\/Arrival time convenient', 'F9': 'Seat comfort', 'F2': 'Inflight entertainment', 'F3': 'Gate location', 'F18': 'Cleanliness', 'F12': 'Ease of Online booking', 'F21': 'Class', 'F4': 'Leg room service', 'F5': 'Age', 'F20': 'Departure Delay in Minutes', 'F10': 'Arrival Delay in Minutes', 'F17': 'Gender', 'F1': 'Checkin service', 'F14': 'Food and drink', 'F7': 'Flight Distance'}

{'F7': 'F6', 'F4': 'F15', 'F2': 'F13', 'F12': 'F19', 'F19': 'F22', 'F17': 'F8', 'F15': 'F16', 'F8': 'F11', 'F13': 'F9', 'F14': 'F2', 'F10': 'F3', 'F20': 'F18', 'F9': 'F12', 'F5': 'F21', 'F16': 'F4', 'F3': 'F5', 'F21': 'F20', 'F22': 'F10', 'F1': 'F17', 'F18': 'F1', 'F11': 'F14', 'F6': 'F7'}

{'C2': 'C1', 'C1': 'C2'}

satisfied

{'C1': 'neutral or dissatisfied', 'C2': 'satisfied'}

BernoulliNB

C1

Used Cars Price-Range Prediction

C1 was the predicted category for the given case and the classifier is shown to be very certain about the above prediction verdict, given that the probability of C1 being the label is about 99.72%. The following five features all contributed positively towards the prediction of the C1 class with increasing levels of impact: F1, F3, F8, F6, and F5. F4 and F7 both had similar levels of impact on the prediction of C1, with F4 having a marginally stronger impact. F4 contributed towards the prediction of C1, while F7 contributed against it, in favour of an alternative label. F10 and F9 are the least relevant features, with very little impact, both with negative attributions, driving the prediction decision or verdict away from C1. From the analysis, only the features, F10, F9, and F7, are shown to have negative attributions, shifting the prediction away from C1. However, the collective attribution of F10, F9, and F7 is very low when compared to that of the positive features, so the classifier is motivated strongly by the positive features, leading to the prediction decision above for the case under consideration.

{'C1': '99.72%', 'C2': '0.28%'}

{'F5': 'Transmission', 'F6': 'Fuel_Type', 'F8': 'Seats', 'F3': 'Name', 'F1': 'Engine', 'F4': 'car_age', 'F7': 'Owner_Type', 'F2': 'Power', 'F9': 'Mileage', 'F10': 'Kilometers_Driven'}

{'F8': 'F5', 'F7': 'F6', 'F10': 'F8', 'F6': 'F3', 'F3': 'F1', 'F5': 'F4', 'F9': 'F7', 'F4': 'F2', 'F2': 'F9', 'F1': 'F10'}

{'C1': 'C1', 'C2': 'C2'}

Low

{'C1': 'Low', 'C2': 'High'}

KNeighborsClassifier

C1

Advertisement Prediction

The ML model or algorithm employed here predicted the class C1 with 100.0% confidence level, clearly implying that the case belongs under the class C1 and not C2 since its associated likelihood is 0.0%. Analysis of the contributions of the features indicated that only features F5 and F1 have negative influence, shifting the classification decision away from C1. However, these features are shown to be the least significant ones when it comes to assigning a label to the case under consideration. Therefore, it is a little surprising to see that the model's confidence level is very high with respect to the prediction made here. Among the remaining positive features, F4, and F7, have the strongest impact or influence, increasing the odds of C1 being the label for the case under consideration and the least positive features are F3, F2, and F6.

{'C1': '100.00%', 'C2': '0.00%'}

{'F4': 'Daily Internet Usage', 'F7': 'Daily Time Spent on Site', 'F6': 'Age', 'F3': 'Area Income', 'F2': 'ad_day', 'F1': 'ad_month', 'F5': 'Gender'}

{'F4': 'F4', 'F1': 'F7', 'F2': 'F6', 'F3': 'F3', 'F7': 'F2', 'F6': 'F1', 'F5': 'F5'}

{'C2': 'C1', 'C1': 'C2'}

Skip

{'C1': 'Skip', 'C2': 'Watch'}

BernoulliNB

C1

Hotel Satisfaction

Judging from the values of the input variables, the label predicted for the case under consideration is C1 with a high confidence level of 98.89%, implying that the probability of C2 being the actual label is just 1.11%. The attribution analysis suggests that F10, F7, and F3 are the most impactful features controlling the label selection. In contrast, F12, F13, and F9 are the least important variables whose values contribute marginally to the label selection. While the variables F10, F13, F11, and F4 contribute towards labelling the given case as C2, the remaining variables such as F7, F3, and F2 strongly support the C1 selection. The variables supporting the assignment of C1 are the positive variables whereas negative variables are those shifting the decision in favour of C2 and are against the C1 labelling decision.

{'C1': '98.89%', 'C2': '1.11%'}

{'F10': 'Type of Travel', 'F7': 'Type Of Booking', 'F3': 'Common Room entertainment', 'F2': 'Stay comfort', 'F1': 'Cleanliness', 'F14': 'Hotel wifi service', 'F4': 'Other service', 'F5': 'Ease of Online booking', 'F8': 'Age', 'F11': 'Checkin\\/Checkout service', 'F15': 'Food and drink', 'F6': 'Departure\\/Arrival convenience', 'F12': 'purpose_of_travel', 'F13': 'Hotel location', 'F9': 'Gender'}

{'F3': 'F10', 'F4': 'F7', 'F12': 'F3', 'F11': 'F2', 'F15': 'F1', 'F6': 'F14', 'F14': 'F4', 'F8': 'F5', 'F5': 'F8', 'F13': 'F11', 'F10': 'F15', 'F7': 'F6', 'F2': 'F12', 'F9': 'F13', 'F1': 'F9'}

{'C2': 'C1', 'C1': 'C2'}

dissatisfied

{'C1': 'dissatisfied', 'C2': 'satisfied'}

RandomForestClassifier

C2

Flight Price-Range Classification

There is little to no doubt that C2, among the three classes, is the proper label for this example since its associated predicted probability is 100.0%. F12, F10, and F7 are the variables with the most influence on the labelling output produced here. Furthermore, these variables have a stronger positive influence on the C2 prediction. Similarly, F11, F5, F6, F8, and F1 are some of the variables favouring the selection of C2 as the correct label. F9, F4, and F2, on the other hand, have a negative and opposing impact on the model, increasing the odds in favour of the other labels. When compared to F7, F12, and F10, all of these negative variables have a moderately low impact on the prediction given here. Finally, the lowest ranked essential input variable is recognised as F3, with a very low positive attribution.

{'C2': '100.00%', 'C1': '0.00%', 'C3': '0.00%'}

{'F10': 'Duration_hours', 'F7': 'Airline', 'F12': 'Total_Stops', 'F11': 'Journey_day', 'F9': 'Source', 'F5': 'Destination', 'F6': 'Journey_month', 'F8': 'Dep_minute', 'F1': 'Arrival_minute', 'F4': 'Arrival_hour', 'F2': 'Duration_mins', 'F3': 'Dep_hour'}

{'F7': 'F10', 'F9': 'F7', 'F12': 'F12', 'F1': 'F11', 'F10': 'F9', 'F11': 'F5', 'F2': 'F6', 'F4': 'F8', 'F6': 'F1', 'F5': 'F4', 'F8': 'F2', 'F3': 'F3'}

{'C3': 'C2', 'C2': 'C1', 'C1': 'C3'}

Low

{'C2': 'Low', 'C1': 'Moderate', 'C3': 'High'}

BernoulliNB

C1

Student Job Placement

Here, the model assigned C1 the highest probability, equal to 99.48%, implying that the predictability of C2 is only 0.52%. Per the attribution analysis, only F2 and F11 have negative contributions that decrease the likelihood of the C1 label in favour of the C2 label. F10, F8, F12, and F5 have the highest positive contributions that improve the odds in favour of the C1. The contributions of the other positive features, such as F9, F3, and F1, have moderate contributions, whilst F4, F6, and F7 are the lowest ranked positive features. All in all, the model is very certain that C2 is not the true label, and this highlighted by the fact that the joint negative contribution of F11 and F2 is only marginal when compared with the very strong influence of positive features such as F10, F12, and F5.

{'C2': '0.52%', 'C1': '99.48%'}

{'F10': 'workex', 'F12': 'specialisation', 'F5': 'ssc_p', 'F8': 'hsc_p', 'F2': 'degree_p', 'F9': 'gender', 'F3': 'degree_t', 'F1': 'etest_p', 'F4': 'hsc_b', 'F11': 'hsc_s', 'F6': 'ssc_b', 'F7': 'mba_p'}

{'F11': 'F10', 'F12': 'F12', 'F1': 'F5', 'F2': 'F8', 'F3': 'F2', 'F6': 'F9', 'F10': 'F3', 'F4': 'F1', 'F8': 'F4', 'F9': 'F11', 'F7': 'F6', 'F5': 'F7'}

{'C2': 'C2', 'C1': 'C1'}

Placed

{'C2': 'Not Placed', 'C1': 'Placed'}

RandomForestClassifier

C1

Employee Attrition

The assigned label or class by the prediction algorithm is C1, which happens to be the most probable class predicted with a probability of around 56.0%, consequently, there is a 44.0% chance that perhaps C2 could be the true label instead. The classification assertion above is attributed to the contributions of mainly F1, F5, F3, F6, F27, F9, F29, F30, F10, F2, F23, F26, F14, F17, F7, F28, F12, F25, F24, and F8. However, not all of the features are considered relevant when determining the correct label for the given case. F22, F16, F21, and F11 are examples of irrelevant features. Among the influential features, F1 and F5 are regarded as the most negative, dragging the verdict in a different direction, while the top features, F3 and F6, have positive contributions, increasing the likelihood that C2 is the right label here. Actually, the reason for the 44.0% prediction likelihood of C2 can be attributed to the strong negative influence of F1 and F5. The other negative features include F27, F9, and F10, while the other positive features are F29, F30, F2, and F23.

{'C2': '44.00%', 'C1': '56.00%'}

{'F1': 'OverTime', 'F5': 'BusinessTravel', 'F3': 'MaritalStatus', 'F6': 'JobInvolvement', 'F27': 'WorkLifeBalance', 'F9': 'Education', 'F29': 'EnvironmentSatisfaction', 'F30': 'Gender', 'F10': 'JobRole', 'F2': 'NumCompaniesWorked', 'F23': 'YearsInCurrentRole', 'F26': 'HourlyRate', 'F14': 'Department', 'F17': 'RelationshipSatisfaction', 'F7': 'PerformanceRating', 'F28': 'YearsWithCurrManager', 'F12': 'Age', 'F25': 'MonthlyRate', 'F8': 'StockOptionLevel', 'F24': 'JobSatisfaction', 'F22': 'DailyRate', 'F11': 'YearsSinceLastPromotion', 'F16': 'YearsAtCompany', 'F21': 'TrainingTimesLastYear', 'F15': 'EducationField', 'F18': 'TotalWorkingYears', 'F20': 'PercentSalaryHike', 'F19': 'MonthlyIncome', 'F13': 'JobLevel', 'F4': 'DistanceFromHome'}

{'F26': 'F1', 'F17': 'F5', 'F25': 'F3', 'F29': 'F6', 'F20': 'F27', 'F27': 'F9', 'F28': 'F29', 'F23': 'F30', 'F24': 'F10', 'F8': 'F2', 'F14': 'F23', 'F4': 'F26', 'F21': 'F14', 'F18': 'F17', 'F19': 'F7', 'F16': 'F28', 'F1': 'F12', 'F7': 'F25', 'F10': 'F8', 'F30': 'F24', 'F2': 'F22', 'F15': 'F11', 'F13': 'F16', 'F12': 'F21', 'F22': 'F15', 'F11': 'F18', 'F9': 'F20', 'F6': 'F19', 'F5': 'F13', 'F3': 'F4'}

{'C1': 'C2', 'C2': 'C1'}

Leave

{'C2': 'Leave', 'C1': 'Leave'}

GradientBoostingClassifier

C1

Broadband Sevice Signup

For the given case, the model predicts C1 as the label. The probability that the label could be the alternative class, C2, is only about 1.94% which implies that the model is very confident in this classification decision or output. F33 and F42 are the top features pushing for the C1 prediction for this case. Other features with a positive impact on this prediction include F31, F10, F38, F12, and F2. On the other hand, the values of F9, F23, F37, and F32 make up the set of features with negative attributions on the prediction decision above. However, compared to F3, F31, F10, and F42, the features above have a very marginal influence on the model. This might explain why the model is highly confident that the true label is likely C1. Finally, there were some features with insignificant impact on the model's prediction decision for the case under consideration and these include F21, F34, F26, and F29.

{'C1': '98.06%', 'C2': '1.94%'}

{'F3': 'X38', 'F42': 'X22', 'F31': 'X32', 'F10': 'X19', 'F38': 'X1', 'F2': 'X13', 'F9': 'X11', 'F12': 'X3', 'F23': 'X16', 'F37': 'X2', 'F32': 'X12', 'F19': 'X14', 'F22': 'X42', 'F4': 'X18', 'F6': 'X28', 'F33': 'X35', 'F18': 'X24', 'F36': 'X20', 'F28': 'X8', 'F27': 'X40', 'F21': 'X34', 'F34': 'X5', 'F29': 'X4', 'F26': 'X41', 'F7': 'X6', 'F14': 'X39', 'F40': 'X7', 'F39': 'X37', 'F24': 'X36', 'F25': 'X33', 'F15': 'X21', 'F5': 'X9', 'F20': 'X31', 'F41': 'X30', 'F11': 'X10', 'F1': 'X27', 'F35': 'X26', 'F17': 'X25', 'F16': 'X15', 'F13': 'X23', 'F30': 'X17', 'F8': 'X29'}

{'F35': 'F3', 'F20': 'F42', 'F29': 'F31', 'F17': 'F10', 'F40': 'F38', 'F11': 'F2', 'F9': 'F9', 'F2': 'F12', 'F14': 'F23', 'F1': 'F37', 'F10': 'F32', 'F12': 'F19', 'F38': 'F22', 'F16': 'F4', 'F26': 'F6', 'F32': 'F33', 'F22': 'F18', 'F18': 'F36', 'F6': 'F28', 'F37': 'F27', 'F31': 'F21', 'F41': 'F34', 'F3': 'F29', 'F39': 'F26', 'F4': 'F7', 'F36': 'F14', 'F5': 'F40', 'F34': 'F39', 'F33': 'F24', 'F30': 'F25', 'F19': 'F15', 'F7': 'F5', 'F28': 'F20', 'F27': 'F41', 'F8': 'F11', 'F25': 'F1', 'F24': 'F35', 'F23': 'F17', 'F13': 'F16', 'F21': 'F13', 'F15': 'F30', 'F42': 'F8'}

{'C1': 'C1', 'C2': 'C2'}

No

{'C1': 'No', 'C2': 'Yes'}

RandomForestClassifier

C1

E-Commerce Shipping

The predicted likelihood of C1 based on the information supplied to the model is 51.62%, whereas there is a 48.38% likelihood that C2 is the correct label. The uncertainty of the model in terms of this case or instance can be attributed mainly to the direction of influence of the variables F6, F2, and F4. Decreasing the chances of C1 being the correct label are the variables F6, F4, F3, and F7. While F6, F4, and F3 have strong negative attributions, F7 is the least negative variable. Increasing the likelihood of C1 prediction are mainly the variables F2, F1, and F10. The features F8, F9, and F5 also have a weak positive influence on the classification decision arrived at by the model for this case under consideration.

{'C1': '51.62%', 'C2': '48.38%'}

{'F6': 'Discount_offered', 'F2': 'Weight_in_gms', 'F4': 'Customer_care_calls', 'F3': 'Product_importance', 'F10': 'Mode_of_Shipment', 'F1': 'Warehouse_block', 'F5': 'Cost_of_the_Product', 'F9': 'Gender', 'F8': 'Customer_rating', 'F7': 'Prior_purchases'}

{'F2': 'F6', 'F3': 'F2', 'F6': 'F4', 'F9': 'F3', 'F5': 'F10', 'F4': 'F1', 'F1': 'F5', 'F10': 'F9', 'F7': 'F8', 'F8': 'F7'}

{'C1': 'C1', 'C2': 'C2'}

On-time

{'C1': 'On-time', 'C2': 'Late'}

SVC

C2

Tic-Tac-Toe Strategy

In this case, the classifier indicates that there is a 99.50% chance that the C2 class is the true label, so it is correct to conclude that the classifier is not sure that C1 is the correct label for the case here. According to the study, five input variables contradict the label choice, while four variables support the classification made above. The variables that contradict the prediction are known as negative features while positive features are those that support the classification verdict. F6, F7, F5, F4, and F2 are the negative variables that reduce the likelihood of C2 being the correct label. F3, F9, F1, and F8 are the positive variables that increase the likelihood of C2.

{'C1': '0.50%', 'C2': '99.50%'}

{'F6': 'middle-middle-square', 'F3': 'top-left-square', 'F9': 'bottom-left-square', 'F1': 'bottom-right-square', 'F7': ' top-right-square', 'F5': 'middle-right-square', 'F4': 'top-middle-square', 'F8': 'middle-left-square', 'F2': 'bottom-middle-square'}

{'F5': 'F6', 'F1': 'F3', 'F7': 'F9', 'F9': 'F1', 'F3': 'F7', 'F6': 'F5', 'F2': 'F4', 'F4': 'F8', 'F8': 'F2'}

{'C2': 'C1', 'C1': 'C2'}

player B win

{'C1': 'player B lose', 'C2': 'player B win'}

LogisticRegression

C1

Employee Promotion Prediction

Considering the values of features such as F10, F11, and F3, the model is very certain (about 99.65% certain) that C1 is the right label for the given case. While F10, F11, and F3 are the most important features, the model paid little attention to F1, F4, and F6 when deciding on the appropriate label here.Overall, driving down the odds of C1 are the negative features F3, F9, F5, and F4, which are shown to support the other label. However, the very high confidence in the above-mentioned decision is chiefly attributed to the positive contributions of F11, F10, F8, F7, and F2.

{'C1': '99.65%', 'C2': '0.35%'}

{'F11': 'avg_training_score', 'F10': 'KPIs_met >80%', 'F3': 'department', 'F8': 'age', 'F7': 'gender', 'F2': 'region', 'F9': 'length_of_service', 'F5': 'recruitment_channel', 'F1': 'previous_year_rating', 'F4': 'no_of_trainings', 'F6': 'education'}

{'F11': 'F11', 'F10': 'F10', 'F1': 'F3', 'F7': 'F8', 'F4': 'F7', 'F2': 'F2', 'F9': 'F9', 'F5': 'F5', 'F8': 'F1', 'F6': 'F4', 'F3': 'F6'}

{'C1': 'C1', 'C2': 'C2'}

Ignore

{'C1': 'Ignore', 'C2': 'Promote'}

KNeighborsClassifier

C1

Suspicious Bidding Identification

With a certainty of 100.0%, the model labels this case as C1 and from the predicted likelihoods across the classes, it can be inferred that the model verdict is that there is a zero chance that the case is under C2. The most significant feature is F8, while the least important attributes are F4, F7, and F9. The moderate features are F2, F3, F1, F6, and F5, ranked in order of their respective attributions on the label predicted. With regards to the direction of influence of each feature, some of the input features have positive attributions in favour of the assigned label and increasing the response of the model in favour of the C1 label, while the remaining ones contradict. F8, F1, F6, and F4 are the positive features, while F2, F3, F5, F7, and F9 are the negative ones, shifting the prediction verdict in the direction of C2.

{'C1': '100.00%', 'C2': '0.00%'}

{'F8': 'Z3', 'F2': 'Z9', 'F3': 'Z4', 'F1': 'Z8', 'F6': 'Z1', 'F5': 'Z5', 'F4': 'Z2', 'F7': 'Z6', 'F9': 'Z7'}

{'F3': 'F8', 'F9': 'F2', 'F4': 'F3', 'F8': 'F1', 'F1': 'F6', 'F5': 'F5', 'F2': 'F4', 'F6': 'F7', 'F7': 'F9'}

{'C2': 'C1', 'C1': 'C2'}

Normal

{'C1': 'Normal', 'C2': 'Suspicious'}

BernoulliNB

C1

Cab Surge Pricing System

The case under consideration can be labelled as either C1 or C3 or C2, and based on values for features such as F9, F7, F2, F1, and F4, the model labelled this test case as C1 with a confidence level equal to 62.29%. However, there is a 28.41% chance that the label could be C3 and a 9.3% chance that it could be C2. All the features used to make the prediction decision have different influences on the model with respect to this test case. That is, while some features positively support the prediction, others have values suggesting any of the alternative labels could be the true label. According to the analysis, F9, F2, F7, and F1 are the top features with the highest impact on the prediction made. The features F9, F2, F1, and F7 are the top attributes positively supporting the prediction of C1. In contrast, F4 and F8 are the features with the most negative attributions, pushing for the prediction of an alternative class. Further decreasing the likelihood of C1 are the features F10, F6, F5, and F3, which all negatively contribute to the model's final decision with respect to the given case. Finally, features F12 and F11 are shown to be less relevant, with positive contributions to the above classification.

{'C1': '62.29%', 'C3': '28.41%', 'C2': '9.30%'}

{'F9': 'Confidence_Life_Style_Index', 'F7': 'Destination_Type', 'F2': 'Customer_Rating', 'F1': 'Type_of_Cab', 'F4': 'Cancellation_Last_1Month', 'F8': 'Trip_Distance', 'F10': 'Var1', 'F6': 'Customer_Since_Months', 'F5': 'Gender', 'F3': 'Var3', 'F12': 'Life_Style_Index', 'F11': 'Var2'}

{'F5': 'F9', 'F6': 'F7', 'F7': 'F2', 'F2': 'F1', 'F8': 'F4', 'F1': 'F8', 'F9': 'F10', 'F3': 'F6', 'F12': 'F5', 'F11': 'F3', 'F4': 'F12', 'F10': 'F11'}

{'C2': 'C1', 'C1': 'C3', 'C3': 'C2'}

C1

{'C1': 'Low', 'C3': 'Medium', 'C2': 'High'}

SVC

C2

Advertisement Prediction

Tasked with labelling a given case as either class C2 or class C1 , the model assigns C2 as the most probable true label, with a confidence level of approximately 99.90%. This confidence level suggests that the probability of C1 being the correct label is only 0.10%. Attribution analysis conducted indicates that all the variables have a different degree of influence or contribution to the model arriving at the above mentioned classification verdict. The features responsible for the very high certainty of the model with respect to the case under consideration are F4, F3, F6, and F5. Actually, the only input variables with a negative contribution also happen to be the least relevant variables, F2 and F1.

{'C2': '99.90%', 'C1': '0.10%'}

{'F4': 'Daily Internet Usage', 'F3': 'Daily Time Spent on Site', 'F6': 'Age', 'F5': 'ad_day', 'F7': 'Area Income', 'F2': 'Gender', 'F1': 'ad_month'}

{'F4': 'F4', 'F1': 'F3', 'F2': 'F6', 'F7': 'F5', 'F3': 'F7', 'F5': 'F2', 'F6': 'F1'}

{'C1': 'C2', 'C2': 'C1'}

Skip

{'C2': 'Skip', 'C1': 'Watch'}

SVC

C2

Water Quality Classification

Even though there is moderately high confidence in the assigned label, the prediction probabilities across the two classes indicate that C1 could be the correct label for this data instance. The variables with primary contributions resulting in the labelling decision above are F5, F2, F9, and F8. As per the attribution analysis, the top two variables, F5 and F2, have a negative impact, influencing the classifier to label the given data as C1 instead of C2. The only other negative variable is F1, with moderate influence compared to the other two negative variables. On the other hand, there are many variables, specifically F9, F8, F7, F4, F3, and F6, that positively support and influence the classifier to assign C2. To a greater degree, the level of uncertainty with respect to this classification instance could be explained away by just looking at the negative variables' fairly strong pull on the classifier towards C1.

{'C1': '38.68%', 'C2': '61.32%'}

{'F5': 'Sulfate', 'F2': 'Hardness', 'F9': 'ph', 'F8': 'Conductivity', 'F1': 'Turbidity', 'F7': 'Chloramines', 'F4': 'Solids', 'F3': 'Trihalomethanes', 'F6': 'Organic_carbon'}

{'F5': 'F5', 'F2': 'F2', 'F1': 'F9', 'F6': 'F8', 'F9': 'F1', 'F4': 'F7', 'F3': 'F4', 'F8': 'F3', 'F7': 'F6'}

{'C1': 'C1', 'C2': 'C2'}

Portable

{'C1': 'Not Portable', 'C2': 'Portable'}

RandomForestClassifier

C2

Student Job Placement

According to the classification model employed here, there is a marginal chance that the true label for this test example is C1. Undoubtedly, the model estimated that the likelihood of the true label being equal to C2 is 99.92%. The above prediction decision is based on the influence of features such as F11, F12, F3, F9, and F1. All these features have significant positive support for the prediction decision here, with the top features being F12 and F3. Furthermore, the features with a moderate influence on the prediction of C2 are F5, F10, F4, and F8. While F8 positively supports labelling the case under consideration as C2, the features F5, F10, and F4 indicate otherwise. Finally, the features with marginal impact are F2, F6, and F7.

{'C1': '0.08%', 'C2': '99.92%'}

{'F12': 'ssc_p', 'F3': 'workex', 'F11': 'hsc_p', 'F9': 'specialisation', 'F1': 'gender', 'F5': 'mba_p', 'F10': 'hsc_s', 'F8': 'ssc_b', 'F4': 'degree_t', 'F2': 'hsc_b', 'F7': 'degree_p', 'F6': 'etest_p'}

{'F1': 'F12', 'F11': 'F3', 'F2': 'F11', 'F12': 'F9', 'F6': 'F1', 'F5': 'F5', 'F9': 'F10', 'F7': 'F8', 'F10': 'F4', 'F8': 'F2', 'F3': 'F7', 'F4': 'F6'}

{'C1': 'C1', 'C2': 'C2'}

Placed

{'C1': 'Not Placed', 'C2': 'Placed'}

SVM

C1

Customer Churn Modelling

Considering the values of the features, the prediction from the model for the case under consideration is C1 and this labelling decision is not 100% certain given that there is a 27.27% probability that it could be C2. For the case under consideration, the assigned label is mainly due to the values of the features F8, F2, F7, and F4 while the least important is F5. The direction of the contributions of the relevant features is summarised in the following sentences: F8 and F2 have a very strong joint positive contribution in favour of class C1 coupled with moderately positive input features F7, F4, and F9, however unlike them, F5 has a very low positive impact on the model for the case here. All of F6, F3, F1, and F10 have a negative impact on the prediction made here, however, their pull is not enough to shift the prediction in the direction of the other class label, C2.

{'C2': '27.27%', 'C1': '72.73%'}

{'F8': 'Age', 'F2': 'IsActiveMember', 'F7': 'Geography', 'F4': 'NumOfProducts', 'F9': 'Gender', 'F6': 'Tenure', 'F3': 'CreditScore', 'F1': 'EstimatedSalary', 'F10': 'Balance', 'F5': 'HasCrCard'}

{'F4': 'F8', 'F9': 'F2', 'F2': 'F7', 'F7': 'F4', 'F3': 'F9', 'F5': 'F6', 'F1': 'F3', 'F10': 'F1', 'F6': 'F10', 'F8': 'F5'}

{'C1': 'C2', 'C2': 'C1'}

Leave

{'C2': 'Stay', 'C1': 'Leave'}

DNN

C1

Concrete Strength Classification

For this case, the classification model's confidence is only about 69.40%, implying that the likelihood of label C2 is about 30.60%. According to the classification attribution analysis, F8 and F2 are the most relevant features, whereas F3 and F5 are the least influential. When the attributions of the features were carefully analysed, only F6, F7, and F1 are identified as negative features since their contributions drive down the prediction likelihood of the assigned label, C1. Conversely, F8, F2, F4, F3, and F5 have a positive influence on the model in support of labelling the given case as C1 instead of C2.

{'C1': '69.40%', 'C2': '30.60%'}

{'F8': 'slag', 'F2': 'water', 'F6': 'cement', 'F7': 'fineaggregate', 'F4': 'flyash', 'F1': 'coarseaggregate', 'F3': 'age_days', 'F5': 'superplasticizer'}

{'F2': 'F8', 'F4': 'F2', 'F1': 'F6', 'F7': 'F7', 'F3': 'F4', 'F6': 'F1', 'F8': 'F3', 'F5': 'F5'}

{'C1': 'C1', 'C2': 'C2'}

Weak

{'C1': 'Weak', 'C2': 'Strong'}

LogisticRegression

C1

Hotel Satisfaction

The model prediction for the test case is C1 and the confidence level of this prediction decision is 91.36%, while the predicted probability of C2 is only 8.64%. According to the attribution analysis, we can see that the features F8 and F1 have negative attributions, pushing the prediction decision towards the alternative label, C2. Conversely, the F11, F10, F6, and F9 have values with a positive impact, shifting the classification decision towards label C1. Furthermore, while the attributes F14 and F7 contradict the prediction made, F13 and F15 have values that support the prediction from the model for the test case under consideration. Finally, F12, F5, F4, and F3 are the least ranked features, and among them, only F3 has a negative influence that contributes marginally to the shift away from labelling the case as C1.

{'C1': '91.36%', 'C2': '8.64%'}

{'F8': 'Type of Travel', 'F1': 'Type Of Booking', 'F11': 'Hotel wifi service', 'F10': 'Common Room entertainment', 'F6': 'Stay comfort', 'F9': 'Other service', 'F14': 'Checkin\\/Checkout service', 'F13': 'Hotel location', 'F7': 'Food and drink', 'F15': 'Cleanliness', 'F2': 'Age', 'F12': 'Departure\\/Arrival convenience', 'F5': 'purpose_of_travel', 'F3': 'Ease of Online booking', 'F4': 'Gender'}

{'F3': 'F8', 'F4': 'F1', 'F6': 'F11', 'F12': 'F10', 'F11': 'F6', 'F14': 'F9', 'F13': 'F14', 'F9': 'F13', 'F10': 'F7', 'F15': 'F15', 'F5': 'F2', 'F7': 'F12', 'F2': 'F5', 'F8': 'F3', 'F1': 'F4'}

{'C2': 'C1', 'C1': 'C2'}

dissatisfied

{'C1': 'dissatisfied', 'C2': 'satisfied'}

RandomForestClassifier

C2

Student Job Placement

In summary, the model predicted an 87.14% likelihood of the class label C2 for the test example under consideration, therefore, there is a chance of about 12.86% that the correct class label could be a different label. The features with the highest impact on the model are F11, F2, F10, and F3, whose values are attributing most to the labeling decision here and among these features, only F3 shows the potential to shift the narrative toward a different label. On impact comparison, features F11, F2, F10 and F3 have higher impact on the model prediction than F5 and F6. Features F11, F2, F10, F5, and F6 show a positive impact shifting towards the prediction of C2. F3 is the most negative of all the set of features passed to the model, F7, F9, and F4 have moderate negative influence, whereas the feature F1 has very little negative impact on the prediction.

{'C2': '87.14%', 'C1': '12.86%'}

{'F11': 'ssc_p', 'F2': 'hsc_p', 'F10': 'degree_p', 'F3': 'workex', 'F5': 'specialisation', 'F6': 'gender', 'F8': 'hsc_s', 'F7': 'etest_p', 'F4': 'degree_t', 'F12': 'mba_p', 'F9': 'ssc_b', 'F1': 'hsc_b'}

{'F1': 'F11', 'F2': 'F2', 'F3': 'F10', 'F11': 'F3', 'F12': 'F5', 'F6': 'F6', 'F9': 'F8', 'F4': 'F7', 'F10': 'F4', 'F5': 'F12', 'F7': 'F9', 'F8': 'F1'}

{'C2': 'C2', 'C1': 'C1'}

Not Placed

{'C2': 'Not Placed', 'C1': 'Placed'}

LogisticRegression

C2

Food Ordering Customer Churn Prediction

Mainly based on the values of the features F37, F32, F28, and F7, the model classifies the given case as C2 with a prediction confidence level of 90.15%. This means that there is only a 9.85% chance that the correct label could be C1. The features that positively contribute to the prediction include F37, F7, F3, and F8, since their influences increase the model's response in favour of assigning the label C2. On the flip side, features dragging the final decision higher towards C1 include F32, F28, F46, and F38, since their values contradict the assigned label here. Finally, the prediction was made with less emphasis on the values of features such as F19, F21, F18, and F15, given that they are shown to have very close to zero influence.

{'C1': '9.85%', 'C2': '90.15%'}

{'F37': 'Unaffordable', 'F32': 'Perference(P2)', 'F28': 'Influence of rating', 'F7': 'Good Food quality', 'F3': 'Delay of delivery person picking up food', 'F8': 'Less Delivery time', 'F46': 'Freshness ', 'F44': 'Politeness', 'F38': 'Ease and convenient', 'F13': 'More restaurant choices', 'F34': 'Missing item', 'F31': 'Order Time', 'F39': 'Gender', 'F24': 'Time saving', 'F16': 'Unavailability', 'F40': 'Late Delivery', 'F41': 'Temperature', 'F6': 'High Quality of package', 'F12': 'Long delivery time', 'F23': 'Poor Hygiene', 'F21': 'Low quantity low time', 'F19': 'Delivery person ability', 'F18': 'Number of calls', 'F15': 'Google Maps Accuracy', 'F11': 'Residence in busy location', 'F25': 'Good Taste ', 'F20': 'Maximum wait time', 'F35': 'Influence of time', 'F17': 'Good Road Condition', 'F33': 'Age', 'F45': 'Order placed by mistake', 'F30': 'Wrong order delivered', 'F1': 'Delay of delivery person getting assigned', 'F22': 'Family size', 'F4': 'Bad past experience', 'F14': 'Health Concern', 'F2': 'Self Cooking', 'F42': 'Good Tracking system', 'F43': 'More Offers and Discount', 'F5': 'Easy Payment option', 'F29': 'Perference(P1)', 'F10': 'Educational Qualifications', 'F9': 'Monthly Income', 'F27': 'Occupation', 'F26': 'Marital Status', 'F36': 'Good Quantity'}

{'F23': 'F37', 'F9': 'F32', 'F38': 'F28', 'F15': 'F7', 'F26': 'F3', 'F39': 'F8', 'F43': 'F46', 'F42': 'F44', 'F10': 'F38', 'F12': 'F13', 'F28': 'F34', 'F31': 'F31', 'F2': 'F39', 'F11': 'F24', 'F22': 'F16', 'F19': 'F40', 'F44': 'F41', 'F40': 'F6', 'F24': 'F12', 'F20': 'F23', 'F36': 'F21', 'F37': 'F19', 'F41': 'F18', 'F34': 'F15', 'F33': 'F11', 'F45': 'F25', 'F32': 'F20', 'F30': 'F35', 'F35': 'F17', 'F1': 'F33', 'F29': 'F45', 'F27': 'F30', 'F25': 'F1', 'F7': 'F22', 'F21': 'F4', 'F18': 'F14', 'F17': 'F2', 'F16': 'F42', 'F14': 'F43', 'F13': 'F5', 'F8': 'F29', 'F6': 'F10', 'F5': 'F9', 'F4': 'F27', 'F3': 'F26', 'F46': 'F36'}

{'C2': 'C1', 'C1': 'C2'}

Go Away

{'C1': 'Return', 'C2': 'Go Away'}

SVC

C2

Bike Sharing Demand

90.58% it the predicted chance that C2 is the correct label for the given case, indicating that the predicted probability of C1 is only 9.42%. Per the feature-attributions, the top-ranked features are F12, F7, and F11, whereas the smallest important or least ranked features are F8, F6, F10, and F3. The influence of intermediate input features like F4, F2, and F9 is considered moderate. The features with positive contributions to the classification above are F11, F2, F8, and F6, while on the other hand, all the remaining features are shown to negatively contribute to the decision above. The main negative features that decrease the probability that C2 is the true label, considering the likelihood of label C1 for this case, are F12, F7, and F4.

{'C2': '90.58%', 'C1': '9.42%'}

{'F12': 'Functioning Day', 'F11': 'Solar Radiation (MJ\\/m2)', 'F7': 'Rainfall(mm)', 'F4': 'Snowfall (cm)', 'F2': 'Hour', 'F9': 'Temperature', 'F1': 'Holiday', 'F5': 'Humidity(%)', 'F8': 'Visibility (10m)', 'F6': 'Dew point temperature', 'F10': 'Seasons', 'F3': 'Wind speed (m\\/s)'}

{'F12': 'F12', 'F7': 'F11', 'F8': 'F7', 'F9': 'F4', 'F1': 'F2', 'F2': 'F9', 'F11': 'F1', 'F3': 'F5', 'F5': 'F8', 'F6': 'F6', 'F10': 'F10', 'F4': 'F3'}

{'C2': 'C2', 'C1': 'C1'}

Less than 500

{'C2': 'Less than 500', 'C1': 'More than 500'}

GradientBoostingClassifier

C2

Paris House Classification

The most likely label for the given scenario, according to this prediction, is C2, which has a prediction probability of 97.02 percent, whereas C1 has a prediction probability of just 2.98 percent. The impact of F5, F3, and F16 is mostly responsible for the aforementioned classification. F2, F6, and F14 are the following groups of features with moderate contributions. F12, F13, F15, and F10, on the other hand, receive minimal attention from the classifier. Given that all four top features have a substantial positive contribution, it's easy to see why the classifier is quite certain that C2 is the correct label in this case. F6, F9, and F8 are also negative features, having a moderate to low influence.

{'C1': '2.98%', 'C2': '97.02%'}

{'F5': 'isNewBuilt', 'F3': 'hasYard', 'F16': 'hasPool', 'F2': 'hasStormProtector', 'F6': 'made', 'F14': 'squareMeters', 'F9': 'floors', 'F8': 'cityCode', 'F4': 'hasGuestRoom', 'F11': 'basement', 'F17': 'numPrevOwners', 'F1': 'price', 'F7': 'numberOfRooms', 'F12': 'garage', 'F13': 'cityPartRange', 'F15': 'hasStorageRoom', 'F10': 'attic'}

{'F3': 'F5', 'F1': 'F3', 'F2': 'F16', 'F4': 'F2', 'F12': 'F6', 'F6': 'F14', 'F8': 'F9', 'F9': 'F8', 'F16': 'F4', 'F13': 'F11', 'F11': 'F17', 'F17': 'F1', 'F7': 'F7', 'F15': 'F12', 'F10': 'F13', 'F5': 'F15', 'F14': 'F10'}

{'C2': 'C1', 'C1': 'C2'}

Luxury

{'C1': 'Basic', 'C2': 'Luxury'}