Datasets:

james-burton

/

aug-text-exps-v3

like 0

LogisticRegression

C1

Used Cars Price-Range Prediction

The prediction model predicts C1 for the case under consideration since the likelihood of C2 which is equal to 30.05%, is lower than that of C1 and this verdict came about mainly based on the values of the input features passed to the model. F5, F2, and F7 are identified as the most influential features with higher impact on the model's labelling decision here and among them F5 and F2 have negative contributions decreasing the model's response towards the assigned label. Furthermore, F7, F9, and F10 have a positive impact on the model and in effect pushes the decision higher towards C1, while F8, F6, and F4 have identical direction of impact as that of F2 and F5. Finally, F1 is the least relevant feature, therefore, its negative attribution has little effect on the model in this case and also the positive influence of F3 further supports the assigned label.

{'C2': '30.05%', 'C1': '69.95%'}

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

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

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

High

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

DecisionTreeClassifier

C1

Credit Risk Classification

The model assigned the label C1 to the given instance since its associated likelihood is far higher than C2. The most relevant features controlling the prediction decision above are F5, F9, and F4. The less relevant ones include F7, F2, and F1. The majority of the features have values, swinging the verdict towards the other class, C2. The only features increasing the likelihood or probability of C1 being the correct label are F5, F8, and F2. Given that only few features positively contribute to arriving at the C1 prediction, it is very strange that the model has 100.0% confidence in its prediction for the selected instance.

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

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

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

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

High

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

RandomForestClassifier

C2

Music Concert Attendance

There is an 80.0% chance that the true label for the given case is C2. Nine out of twenty features have a positive impact. Most features have a moderately low positive or negative impact, with the exception of F20, F5, and F7 and it appears as if F20 has an extremely negative impact, while F5 and F7 have the greater positive impacts. F18 has positive impacts, whereas the attributions of the features F1 and F12 are negatives. The least important features include F4, F14, F16, F3, F15, F11, and F8 with varying smaller effects.

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

{'F20': 'X11', 'F5': 'X1', 'F7': 'X6', 'F1': 'X10', 'F18': 'X14', 'F12': 'X16', 'F19': 'X13', 'F17': 'X12', 'F10': 'X3', 'F4': 'X2', 'F14': 'X15', 'F16': 'X4', 'F3': 'X7', 'F15': 'X17', 'F11': 'X8', 'F8': 'X5', 'F13': 'X18', 'F6': 'X19', 'F9': 'X9', 'F2': 'X20'}

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

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

< 10k

{'C2': '< 10k', 'C1': '> 10k'}

RandomForestClassifier

C2

Employee Attrition

There is disagreement about which label is acceptable for the case under consideration since the model is unsure which of the two labels is right. The confusion in the aforementioned classification may be attributable only to the effect of F1. F1 is by far the most influential variable, with a negative contribution that reduces the chance of label C2 being the correct label in the given case substantially; supporting the that case should be labelled as C1. Compared to the influence of F1, the remaining variables have a moderate to low effect on the classification decision made here for the case under consideration. F16, F25, and F23 are notable moderately key variables, with positive contributions boosting the likelihood of label C2. F22, F12, F29, F28, F20, F6, F7, F27, F24, and F10 are not among the features demonstrated to contribute to the classification above; since they have very insignificant impact on the model's conclusion here.

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

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

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

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

Stay

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

KNeighborsClassifier

C2

German Credit Evaluation

For the case under consideration, the probability of C1 being the correct label is only 12.50%, implying that there is an 87.50% chance that C2 is the true label. The decision above was arrived at mainly based on the values of the following variables F6, F8, and F2. Among these top variables, only F6 has a very strong positive impact on the model, increasing the likelihood of C2 prediction. The most important variables decreasing the prediction are F8 and F2 and the remaining two shifting the verdict away from C2 are F7 and F5. F1 and F9 are the lowest-ranked variables, less important to the prediction made here since they have a moderately low positive impact on the model.

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

{'F6': 'Checking account', 'F8': 'Saving accounts', 'F2': 'Purpose', 'F7': 'Sex', 'F5': 'Duration', 'F4': 'Housing', 'F3': 'Age', 'F1': 'Job', 'F9': 'Credit amount'}

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

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

Good Credit

{'C2': 'Good Credit', 'C1': 'Bad Credit'}

DecisionTreeClassifier

C2

Hotel Satisfaction

With a high degree of confidence, close to 100 percent, the classifier's final label choice for the given case is C2 due to the predicted probability distribution between the class labels. Analysis of the attributions of the input features indicates that the most relevant features driving the classification above are F10, F2, F5, and F7, whereas F11 and F6 are shown to have little contribution to the decision. Furthermore, only four of the features have a negative influence, swinging the classifier decision in this case towards the C1 label and they are F2, F9, F11, and F6. However, except for F2, the contribution of the other negative features is very low when compared with the top positive features such as F5, F13, and F7.

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

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

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

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

satisfied

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

KNNClassifier

C2

Car Acceptability Valuation

Based on the values of the six input features, the model assigned the label C2 to the given case with a higher degree of confidence and according to the model used here, there is a near-zero chance that the label could be C1. Influencing the prediction assessment above are the top four features, F4, F3, and F1, whereas, the least significant feature here is F2. Among the input features, only two, F3 and F2, contradict the label assignment decision above since their values are shifting the label decision in the C1 direction. However, the joint attribution of these features is outweighed by the remaining four features, F4, F1, F5, and F6. This could explain why the model is very certain about the C2 prediction made for the case under consideration.

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

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

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

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

Acceptable

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

RandomForestClassifier

C1

Printer Sales

The most likely label for the given data is C1 and this decision is as the result of the variables passed to the classifier. F12, F7, F20, and F13 are the primary contributors to the aforementioned prediction output. F18, F22, F15, F2, F24, and F17, on the other hand, make insignificant contributions to the classifier labelling the given example. F4 and F26, as well as F8, F16, have a moderate influence on the label selection. The classifier's confidence in the label decision above might be explained away by comparing the greater positive attributions of F26, F4, F12, and F13 to the negative attributions of F8, F23, F20, F9, F7, and F14.

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

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

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

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

More

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

LogisticRegression

C1

Flight Price-Range Classification

Because the prediction algorithm outputs reveal that the likelihood of C1 being the correct label is equal to 93.02%; hence, there is only a little possibility that the true label for the provided data instance is either of the other labels, C4, C3, and C2. The variables F1, F11, F12, and F7 are the most crucial ones driving the label assignment conclusion above, whereas F4, F3, and F8 are the least vital ones. Taking into account the direction of effect of each input feature, as demonstrated by the attribution analysis, it is possible to deduce that the positive features driving the prediction upward towards C1 are F1, F9, F12, F5, F7, F11, and F3. The negative contributions of F2, F6, F8, F4, and F10 are ascribed to the marginal uncertainty in the expected output decision. When the predicted probabilities across the classes are considered, it is possible to infer that the combined positive contribution outranks the negative contributions; therefore, the algorithm is certain that C1 is the real label.

{'C1': '93.02%', 'C3': '6.97%', 'C2': '0.01%', 'C4': '0.0%'}

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

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

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

Low

{'C1': 'Low', 'C3': 'Moderate', 'C2': 'High', 'C4': 'Special'}

LogisticRegression

C2

Food Ordering Customer Churn Prediction

Based mainly on the values of the input variables F13, F11, F7, and F38, the predictor classifies the case as C2 with a 90.15% labelling confidence level, indicating that there is only a 9.85% probability that the right label could be C1. Variables that contribute positively to the prediction verdict include F13, F10, F44, and F38. The values of these variables increase the odds of the model labelling the given case as C2. On the other hand, F7, F11, F32, and F17 are the variables influencing the prediction decision in favour of C1 instead of C2. Simply put, the values of these negative variables contradict the label assigned here and finally, the model places little emphasis on the values of features such as F41, F5, F45, and F30 when determining the correct label in this instance, as they have nearly zero influence.

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

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

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

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

Go Away

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

LogisticRegression

C1

Air Quality Prediction

The classification output decision is based solely on the information supplied to the model and it predicts class C1 with a higher confidence level, equal to 94.10%, indicating the model is very confident that the correct label for the given case is not either class C3 or class C2 or class C4. The classification output decision with regards to the given case boils down to the values of the features F1, F4, F3, and F2, which are shown to have the most significant influence on the model. Among these relevant features, only F1, F2, and F4 have a positive impact, increasing the response towards labelling the case as C1. Conversely, the remaining ones, F3 and F6, have negative attributions, decreasing the odds of the assigned label. Finally, feature F5 has little impact on this prediction among the features since its value received little consideration from the model.

{'C3': '0.00%', 'C2': '0.53%', 'C1': '94.10%', 'C4': '5.37%'}

{'F1': 'MQ5', 'F4': 'MQ6', 'F2': 'MQ3', 'F3': 'MQ4', 'F6': 'MQ1', 'F5': 'MQ2'}

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

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

Cleaning

{'C3': 'Preparing meals', 'C2': 'Presence of smoke', 'C1': 'Cleaning', 'C4': 'Other'}

SVC

C2

German Credit Evaluation

The model predicts that this case is likely C2 with a confidence level equal to 66.80%, meaning there is a 33.20% chance that it could be C1 instead. According to the analysis for this case under consideration, the most relevant features considered by the model are F7, F6, F9, F8, and F5, however, the least relevant features are F2 and F3. The F9, F8, F5, and F4 can be regarded as positively supporting features given that they increase the model's response in favour of the prediction conclusion above. In contrast, the F7, F6, and F1 are the features supporting the prediction of the alternative or other class label C1. Even though only a small number of features support the prediction of C1, their collective or joint influence is enough to upset the joint influence of the other features, leading to the uncertainty of the C2 prediction.

{'C2': '66.80%', 'C1': '33.20%'}

{'F7': 'Saving accounts', 'F6': 'Duration', 'F9': 'Checking account', 'F8': 'Sex', 'F5': 'Age', 'F4': 'Purpose', 'F1': 'Housing', 'F2': 'Job', 'F3': 'Credit amount'}

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

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

Good Credit

{'C2': 'Good Credit', 'C1': 'Bad Credit'}

SGDClassifier

C2

House Price Classification

Based on the values of the input variables resulting in the predicted likelihoods across the classes, the classification algorithm is confident that the right label for the provided data is C2. According to the algorithm, there is no possibility that C1 is the correct label. However, the attributions of F1, F12, F5, and F8 indicate that the correct label might be C1 rather than C2. The top four variables are F13, F4, F2, and F9, all of which have a positive influence on the algorithm's prediction output, hence confirming the C2 classification. This conclusion is further supported by the contributions of F11, F10, F3, F7, and F6, which are also positive variables.

{'C1': '0.0%', 'C2': '100.0%'}

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

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

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

High

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

KNeighborsClassifier

C1

Suspicious Bidding Identification

With a higher degree of confidence, the classifier assigns the label C1 due to the fact that there is a close to zero chance that C2 is the label. The confidence level with respect to this classification output is largely due to the strong positive influence of F7. However, decreasing the probability that C1 is the true label are the negative features F8, F1, F4, F6, F3, and F9. Furthermore, F2 and F5 also increase the likelihood of C1 being the true label. In conclusion, the joint impact of the negative features is very weak compared to the positive features, hence the strong driving force of the classifier to assign the chosen label, C1.

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

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

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

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

Normal

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

MLPClassifier

C1

Annual Income Earnings

According to the input variables, there is a 99.81% chance that C1 is the correct label for the given data instance, with a prediction probability of the alternative label, C2, equal to 0.19% which shows that there is little chance that C2 is the true label. F13, F8, and F12 are the top contributing features leading to the classification decision here. On the contrary, the F5, F14, and F4 are the least relevant features. The input features with moderate influence are F6, F3, F7, F10, F1, F9, and F2. Even though the different features have some level of influence on the classification, not all of them positively contribute. Actually, F7, F11, F2, and F14 have negative attributions, decreasing the classifier's response towards assigning C1; however, the joint influence of these features is outweighed by the positive attributions of F13, F8, F12, F6, and F3.

{'C1': '99.81%', 'C2': '0.19%'}

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

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

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

Under 50K

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

SVC

C2

Australian Credit Approval

With a confidence level equal to 81.43%, the classification algorithm labels the given data as C2, however, there is about an 18.57% chance that C1 could be the right label. The assignment of C2 to the given case is mainly based on the positive influence and contribution of input features F13, F6, and F2. Furthermore, the majority of the remaining input features have positive contributions, further increasing the predictability of label C2. F8, F9, F3, and F14 are the features with negative contributions, shifting the decision towards C1 instead of C2. Summarizing, comparing the attributions of the negative features to even those of the top three positive features explains why the algorithm is certain that C2 is the right label here.

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

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

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

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

Class 2

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

DNN

C2

Broadband Sevice Signup

The classification model employed here is very certain that the correct label is C2, implying that there is a near-zero chance that C1 is the label. The top six variables with the most influence on the prediction are all shifting the prediction in favour of C2. This might explain why the model is very certain about the C2 label and these top positive attributes are F5, F36, F2, F14, F19, and F35. F33, F24, F17, F38, and F20 all have moderately low-negative contributions, weakly swinging the direction of the model's decision towards C1. Finally, the decision to label the case as C2 is marginally supported by F18 and F8, whereas F41, F3, and F30 suggest that C1 could be the true label. In conclusion, the very high confidence level with regard to this prediction can be explained away by considering the very strong positive influence of F5, F36, F14, and F2.

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

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

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

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

Yes

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

KNeighborsClassifier

C2

Water Quality Classification

For the given case, the model generates the label C2 instead of C1, since C2 has a higher prediction likelihood than C1. According to the attribution graph shown, F8, and F7 are the most influential variables, resulting in the classification verdict above. F4, F3, and F5, on the other hand, are the least important variables considered by the model. F6, F2, F1, and F9 are shown to have a moderate influence on the classification made here. To sum up, with F4, F3, and F5 being the only variables contributing negatively, it is foreseeable why the model is quite certain that C1 is not the correct label for the given case.

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

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

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

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

Not Portable

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

DecisionTreeClassifier

C2

Concrete Strength Classification

Per the classifier, the most probable class with a very high confidence level is C2 mainly because the probability that C1 is the correct label is zero. From the attributions analysis, all the inputs are shown to contribute to or influence the above classification. The ranking of the features from the least important to the most important based on their degree of influence is as follows: F2, F6, F1, F3, F7, F4, F5, F8. Simply looking at the attributions of the input features, it is obvious why the classifier is very confident that C1 is not the correct label for the given All the features have positive contributions, resulting in a strong push towards C2.

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

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

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

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

Weak

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

KNeighborsClassifier

C1

Air Quality Prediction

The model predicted the C1 class for the test case with a very high degree of confidence. F3 is the only feature contributing against the prediction of the C1 class, while F6 and F5 contributed positively towards the prediction of C1. In decreasing order, F1, F4 and F2 were the three features with the least positive impact on the prediction of C1. Overall, given that only F3 has negative influence on the decision, it is not surprising to see the associated confidence level of the assigned label.

{'C2': '0.00%', 'C1': '100.00%', 'C4': '0.00%', 'C3': '0.00%'}

{'F3': 'MQ6', 'F6': 'MQ4', 'F5': 'MQ5', 'F1': 'MQ2', 'F4': 'MQ1', 'F2': 'MQ3'}

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

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

Presence of smoke

{'C2': 'Preparing meals', 'C1': 'Presence of smoke', 'C4': 'Cleaning', 'C3': 'Other'}

SVM_linear

C2

Mobile Price-Range Classification

Per the classification algorithm, the most probable class is C2 since the prediction probabilities indicate there is little to no chance that the correct label for the given data instance is any of the following classes: C1, C3, and C4. This labelling is primarily owing to the roles that the features F16, F9, and F3 performed. On the lower end of the spectrum are the input features F13, F6, F15, and F19, which are demonstrated to be less essential for this labelling assignment task. Finally, only F17 and F2 are features having a negative effect, reducing the likelihood of C2 being the accurate classification here.

{'C1': '0.00%', 'C3': '0.00%', 'C4': '0.00%', 'C2': '100.00%'}

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

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

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

r4

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

GradientBoostingClassifier

C1

Australian Credit Approval

The predicted label is C1 at a confidence level of 92.11%, insinuating that there is a 7.89% chance that the label could be C2. In this case, the feature with the most significant influence on the model's decision is F8, with a very strong positive contribution in support of the C1 prediction. The next set of features with moderately high impact is F3, F7, F5, F11, and F12. Among this set, only F12 and F3 have a negative influence in support of label C2. Finally, on the lower end, the values of F1, F9, and F2 are deemed less important by the model when labelling this case.

{'C2': '7.89%', 'C1': '92.11%'}

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

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

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

Class 2

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

SVM

C2

Customer Churn Modelling

Taking into account the values of the input features, the prediction model's output for the case under consideration is C2. Given that there is a 27.27% probability that it could be C1, this labelling decision is not 100.0% certain. For the case under consideration, the label assignment is mainly due to the values of F3, F10, F1, and F5. F5 is identified as the most important or relevant, while F4 is considered the least important, since its contribution to the model is only marginal. In terms of the influence direction of each feature, F5 and F10 have a very strong positive contribution, driving the prediction higher toward the C2 class followed by F3, F1, and F6 all with moderately positive influence, whereas F4 has a negligible positive impact on the model in this case. Finally, for this case, F8, F2, F9, and F7 all have a negative impact on the prediction verdict, however, their pull or influence is not enough to transfer predictions in the direction of another class label, C1.

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

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

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

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

Leave

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

SVMClassifier_liner

C1

Employee Attrition

The prediction output decision by the model is that the likelihood of label C1 is 94.15% and that of class C2 is only around 5.85%, meaning the model is certain that C1 is likely the true label for the given case. First of all, the classification is performed with negligible contributions from the variables F4, F9, F13, F21, and F24 since their attributions are very close to zero. However, examination or inspection of the attributions of the different variables reveals that F8, F18, F17, F16, and F30 are the highly influential ones driving the predicted probabilities across the classes. In addition, the decision about the correct label for this case is moderately influenced by the values of F19, F15, F20, F7, F12, F10, and F6. In terms of the direction of influence or contributions of the variables, F8, F17, F16, F15, and F20 are the top positive variables, encouraging the predicted output to be equal to C1. Pushing the decision towards the C1 label and further away from C2 are the contriutions of the variables F12, F10, F22, and F14. Finally, the 5.85% likelihood of C2 can be attributed to the negative contributions of the top negative variables F18, F30, F14, F7, and F6.

{'C1': '94.15%', 'C2': '5.85%'}

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

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

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

Stay

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

RandomForestClassifier

C2

Printer Sales

The most probable label, according to the classifier for the given data, is C2, which happens to have a higher predicted probability than that of C1. The major players in the above prediction output are F5, F22, F6, and F7. Conversely, F3, F26, F8, F13, F18, and F17 have negligible contributions when it comes to the classifier labelling the given case. Features such as F16, F10, F20, and F21 have a moderate influence on the decision. Comparing the stronger positive attributions of F22, F5, F10, and F20 to the negative attributions of F6, F7, F16, F14, F2, and F23 could explain why the classifier is quite confident in the label choice above.

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

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

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

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

More

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

AdaBoostClassifier

C4

Air Quality Prediction

The most likely label is C4 since there is a 30.83% chance it could be C2, a 35.74% chance it could be C4, and a 33.42% chance it could be C3. Therefore, the correct label is not C1, which the model is very certain about. The above decision is primarily controlled by the values F6, F2, F5, and F1 which are shown to have positive influences that support the model's classification judgement here. In contrast, the remaining features F3 and F4 negatively support the classification decision, decreasing the chances of C4 being the correct label. In view of the fact that the probability distributions across the classes, we can conclude that the model is very uncertain about which label is appropriate for the given data instance and the features F3 and F4 should be blamed for this.

{'C2': '30.83%', 'C4': '35.74%', 'C1': '0.00%', 'C3': '33.42%'}

{'F6': 'MQ5', 'F2': 'MQ3', 'F5': 'MQ2', 'F1': 'MQ6', 'F3': 'MQ1', 'F4': 'MQ4'}

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

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

Presence of smoke

{'C2': 'Preparing meals', 'C4': 'Presence of smoke', 'C1': 'Cleaning', 'C3': 'Other'}

KNeighborsClassifier

C1

Ethereum Fraud Detection

Because the prediction probability of C2 is equal to 0.0%, the presented case is labelled as C1 with a very high level of confidence. For this classification scenario, the input features that have the greatest influence on the end outcome are F20, F7, F2, and F6. F19, F12, F8, F21, and F11 have a mild impact. However, because F14, F24, F5, and F1 have insignificant attribution values, they have little influence on the model's judgement. Among the top features, F20, F7, F2, and F6, only F20 and F2 exhibit negative attributions that favour the least likely class, C2, whereas F7 and F6 positively support the model's classification result for the provided data. Finally, only F32 and F29 positively contribute to the model's decision among the remaining significant features: F29, F36, F33, F32, F3, F9, and F26.

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

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

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

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

Fraud

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

RandomForestClassifier

C1

Mobile Price-Range Classification

Between the four possible classes, the label for this case is predicted as C1, with a 73.08% likelihood that this is correct. With a likelihood of about 26.92%, the next probable label is shown to be C2. The prediction assessment above is mainly based on the values of the features F2, F3, F1, F14, and F10. The strongest impact came from F2, followed by F1, F3, F10, and F14. The collective contributions of the positive features F2, F3, F11, and F15 far outweigh the contributions of the negative attributes F1, F10, F14, and F4. Of the twenty attributes, majority of them are shown to have values pushing the prediction towards one of the three other possible classes and as such, it is surprising to see that the model is not 100% confident in the C1 prediction. On the grounds that the likelihood of C1 being correct is 73.08%, we can conclude that the model is quite confident with its final decision for the case under consideration.

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

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

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

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

r1

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

SVC

C2

Health Care Services Satisfaction Prediction

The prediction probability associated with class C1 and class C2, respectively, is 35.34% and 64.66%. Based on these probabilities, the model labels the given case as C2 since it is the most probable class. According to the attribution analysis, the most relevant features considered by the model here are F12, F2, and F6, while the least relevant features are F14, F5, and F8. Regarding the direction of influence of the features, F12, F2, F6, and F15 are the top positively supporting features, driving the decision higher in favour of C2. Further increasing the probability that C2 is the true label are the values of other positive features such as F9, F16, F4, and F1. To explain why the likelihood of C1 is 35.34%, we have to look at the negative contributions from F10, F13, F7, F5, F14, and F8. The abovementioned negative features contradict the model's decision with respect to the classification outcome.

{'C1': '35.34%', 'C2': '64.66%'}

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

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

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

Satisfied

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

KNeighborsClassifier

C1

Real Estate Investment

The classifier is very uncertain about the correct label for the case given. Regarding the classifier's decision, there is close to an even split on the probability of either of the possible labels is the correct label but the classifier chooses the label as C1. The prediction verdict above is attributed to the contributions of mainly the following features: F11, F2, F5, and F13, however, the lowest ranked features are F16, F3, and F18. Analysing the direction of influence of the features shows that there are ten positive and ten negative features. Positive features such as F5, F13, F14, and F4 increase the response of the classifier in favour of the assigned label. Conversely, negative features such as F11, F2, F6, and F8 decrease the likelihood of C1 being the correct label given that their values support the alternative label, C2. The uncertainty concerning the label assignment can be due to the fact that the top negative features F11 and F2 have very high attributions shifting the classifier's verdict away from the C1 class.

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

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

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

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

Ignore

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

DecisionTreeClassifier

C1

Car Acceptability Valuation

The classification algorithm believes that C1 is the output label that was generated with 100% certainty and that C2 is unlikely to be the correct label in this case. According to the attribution investigations, the following input features are ranked from most relevant to least relevant: F5, F3, F4, F2, F1, and F6. As shown by the attribution plot, F5 is the only one shown to positively contribute to the above classification decision, while the others contribute negatively. The contributions of negative features such as F3, F4, and F2 result in the decision being driven in a different direction. From the prediction confidence level, we can conclude that the very strong influence of F5 overshadows the contributions of the negative features hence the very high confidence level.

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

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

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

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

Unacceptable

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

RandomForestClassifier

C1

Flight Price-Range Classification

Of the three possible labels, there is 100.0% confidence that C1 is the most probable label for the given case. The features that heavily influence the classification verdict presented here are F5, F4, and F1, and they have a very strong positive contribution, increasing the odds of the C1 prediction. Other features with a positive influence on the model are F7, F12, F2, F9, and F10. On the contrary, F6, F3, and F8 make the model's decision fluctuate negatively towards selecting an alternative label. All of the negative features mentioned above have a low to moderate impact on the classification verdict presented here compared to F1, F4, and F5. Finally, F11 with its very low positive impact is the least ranked feature marginally pushing the decision towards the assigned label.

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

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

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

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

Low

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

RandomForestClassifier

C2

Used Cars Price-Range Prediction

Per the model, class C1 has a prediction probability of 10.50 percent, whereas class C2 has a predicted probability of 89.50 percent. As a result of the model, it can be determined that C2 is the most likely label for the given scenario. All of the input features are shown to contribute to the above conclusion, with F9, F1, and F2 having the most influence on the classification decision. The least influential features with regard to this classification are F10, F4, F6, and F5, whereas, the impact of F7, F3, and F8 can be classified as modest. The large positive contributions of F1 and F9 are responsible for the model's high confidence which further supported by the positive contributions of F7, F10, and F4. In conclusion, the negative features F2, F3, F6, F8, and F5 favour labelling the case as C1 hence the associated predicted probability.

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

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

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

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

High

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

AdaBoostClassifier

C1

Basketball Players Career Length Prediction

The classifier says that C1 is the most likely label for the provided data with relatively high confidence. It is crucial to remember, however, that there is a 21.80% possibility that it is C2. F3 and F11 are the major driving variables for the aforementioned classification or prediction choice. The remaining variables F18, F12, F13, and F19 have a modest to minor impact on the selection made above. Among the input variables, F18, F19, F6, F15, and F16 are the subset that have a negative influence or contribution whereas all of the remaining variables have a positive impact. In essence, the substantial positive contributions of F3 and F11, together with the contributions of additional positive variables such as F12, F13, F5, and F2, account for the classifier's confidence in this classification.

{'C1': '78.20%', 'C2': '21.80%'}

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

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

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

More than 5

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

RandomForestClassifier

C2

Printer Sales

C2 has an 83.0% chance of being the correct label for the case under consideration, making C1 the least likely class with a predicted likelihood of 17.0%. F2, F8, and F9 features have a significant impact on class selection here while on the other hand, the remaining features are shown to have marginal to no contribution to the classification verdict here. In actual fact, the values for F15, F23, F4, F21, F5, and F6 may have been ignored by the classifier because their respective influences are almost zero. Of the important features, only F13, F3, F12, F22, F1, and F7 are negative and this is mainly because their contribution to selection tends to reduce the chance that C2 is the correct label, preferring that the case is classified as C1. The remaining features such as F2, F8, F9, F10, F19, and F14 strongly contribute positively, increasing the chances of C2 which explains the level of certainty associated with C2.

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

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

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

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

Less

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

LogisticRegression

C2

Music Concert Attendance

C2 is the label picked by the algorithm with about 82.06% certainty, since the prediction likelihood of C1 is only 17.94%. F7, F10, F14, and F18 all contribute significantly to the above classification output and among them, the features that support the most positive contribution to the C2 prediction are F18, F7, and F10, while F14 drives the final prediction against assigning C2 in support of C1. F13 also contributes positively to the classification here, but F17 contributes negatively and like F14 favours C1. Finally, according to the analysis, F20, F19, F1, and F9 all have little effect on the final prediction made by the algorithm for this case.

{'C1': '17.94%', 'C2': '82.06%'}

{'F18': 'X11', 'F7': 'X1', 'F14': 'X13', 'F10': 'X3', 'F17': 'X8', 'F13': 'X6', 'F2': 'X2', 'F6': 'X9', 'F5': 'X17', 'F15': 'X10', 'F3': 'X4', 'F12': 'X14', 'F4': 'X20', 'F8': 'X18', 'F11': 'X19', 'F16': 'X7', 'F1': 'X12', 'F20': 'X15', 'F9': 'X16', 'F19': 'X5'}

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

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

> 10k

{'C1': '< 10k', 'C2': '> 10k'}

LogisticRegression

C3

Flight Price-Range Classification

The model is confident in its prediction, as it predicted class C3 with a likelihood of 90.48% and hence, for the given case, there is a smaller chance of it being any other class label. F1 and F4 are deemed the most important features whereas on the other hand all the other features have moderate to minimal amounts of influence. Both F1 and F4 have the same direction of impact, increasing the odds of the predicted label, C3. While F6 and F11 are both encouraging the model to make a prediction of C3, the others F8, F12, and F5 is pushing the model towards a different label. Many features have moderately low impact on the final prediction, but the features F3, F5, and F10 are those with the smallest influence.

{'C3': '90.48%', 'C2': '9.51%', 'C1': '0.01%'}

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

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

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

Low

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

DecisionTreeClassifier

C1

Airline Passenger Satisfaction

Based on the probability distribution across the classes, the classifier is shown to have a moderately high confidence level in the C1 label assignment, with its likelihood equal to 65.0%, whereas that of C2 is only 35.0%. The prediction decision above is predominantly due to the influence of the variables F14, F4, F2, and F3. On the lower end are the least relevant variables, F5, F26, F11, F19, F24, and F21, with little to no influence on the classifier when assigning a label to the given instance. On the one hand, the top positive variables are F14, F4, and F2, increasing the probability that C1 is the correct label. Also, the top negative variables are F3, F6, F9, and F23, decreasing the classifier's response and consequently shifting the prediction verdict in the opposite direction towards C2. Other variables with a positive direction of influence are F15, F7, F17, F10, F25, F8, F1, F18, and F22.

{'C2': '35.00%', 'C1': '65.00%'}

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

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

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

Acceptable

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

KNeighborsClassifier

C1

Company Bankruptcy Prediction

The model's output labelling judgement for the case under consideration is as follows: C2 cannot be the label for the given case; C1 is the most likely class label with a 100.0% confidence level. The key driving factors resulting in the aforementioned classification are the values of the input features: F27, F7, F70, F66, F54, F19, and F33. F89, F67, F46, F5, F87, F80, F17, F52, F41, F39, F23, F32, and F75 are the features that have a modest effect on the decision. Aside from the aforementioned input features, all others, such as F11, F13, F55, and F92, are revealed to be irrelevant to the conclusion reached here. Not all of the influential features support labelling the current instance as C1, and they are referred to as negative features. F33, F67, F23, F32, and F75 are the negative attributes that diminish the likelihood that C1 is the correct label in this case. F27, F7, F70, and F66 are important positive features that strongly increase the likelihood that C1 is the correct label.

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

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

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

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

No

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

BernoulliNB

C2

Student Job Placement

For the case under consideration, the model assigned C2 with very high confidence, since the likelihood of C1 being the right label is only 0.52% which is very small. F7, F5, F10, and F3 have a large positive impact on the model's output prediction. F10 and F3 have a moderately positive impact on the prediction of C2, while F6 has a similar impact but in the opposite direction. F11, F9, and F8 have a very low impact on classification. F12, F4, F1, and F2 have a larger but still insignificant effect. Examining the attributions indicates that there are only two features, F6 and F9, with values that contradict the prediction made here but, their impact on the model is smaller when compared to positive features such as F5, F10, and F7, which explains why the confidence level associated with this classification is high.

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

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

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

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

Placed

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

LogisticRegression

C1

Flight Price-Range Classification

Since the likelihood of C1 being the true label is shown by the prediction algorithm outputs to be equal to 93.02 percent, there is only a small chance that the true label for the given data instance is any of the other class labels, C3 and C2. The features F9, F4, F7, and F2 are the most important ones driving the label assignment verdict above, and on the other hand, the least relevant features are shown to be F5, F1, and F11. Considering the direction of influence of each input feature, as shown by the attribution analysis, it can be concluded that the positive features steering the prediction higher towards C1 are F9, F4, F2, F7, F10, F8, and F1. The marginal doubt in the predicted output decision is attributed to the negative contributions of F3, F6, F11, F5, and F12. Considering the attributions of the features and predicted probabilities across the classes, it can be concluded that the joint positive contribution outranks the negative contributions; hence, the algorithm is confident that C1 is likely the true label.

{'C1': '93.02%', 'C3': '6.97%', 'C2': '0.01%'}

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

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

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

Low

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

RandomForestClassifier

C2

Cab Surge Pricing System

Between the three possible classes, there is an 88.0% probability that the correct label for this case is C2. This means that there is a 12.0% chance that the label could be one of the other possible labels, C1 or C3. Increasing the odds of the predicted label are the variables F2, F5, F1, and F6. The next set of variables, F10, F3, and F7, have values that moderately decrease the likelihood of C2 being the correct label. F9, F4, and F12 are the other negatively contributing features, and given that they are lowly ranked, they have a marginal impact when determining the correct label for this case. The other positive features further increasing the probability that C2 is the right label are F11 and F8. Overall, we can conclude that the decision to label the case as C2 is largely due to the strong positive influence of F5, F2, F6, and F1.

{'C1': '3.00%', 'C3': '9.00%', 'C2': '88.00%'}

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

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

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

C3

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

RandomForestClassifier

C1

Wine Quality Prediction

Based on the input variables, the model is moderately confident that the C1 is the appropriate label for the data under consideration. As a matter of fact, the prediction likelihood associated with class C2 is about 30.42%. The preceeding classification verdict can be largely blamed on the contributions of variables F7, F3, F10, and F5, whereas those with marginally lower contributions are F9, F1, and F4. The variables with moderate contributions are F6, F11, F8, and F2. Considering their respective contributions, F7, F10, F5, and F2 are the variables with positive influence that increase the chances of C1 being the correct label for the given data. The little doubt in the label choice here could be attributed to the negative variables, mainly F3, F6, F8, and F11, which decrease the chances of the model labelling the data given as C1 since these negative variables favour selecting the alternative label, C2 over C1. Given that majority of top variables contribute positively, it is not unexpected that C1 is the picked label with reasonably high confidence.

{'C2': '30.42%', 'C1': '69.58%'}

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

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

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

high quality

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

SVC

C2

E-Commerce Shipping

The classifier is 69.02% certain that the given case is under the class label C2, implying that the likelihood of C1 is only 30.98%. Analysis performed to understand the contribution of each input feature revealed that: F3, F6, and F9 are the most influential features when assigning a label to the given case. Features F1, F4, F8, and F2 have moderate contributions, whereas the F5, F7 and F10 have lower relevance to the final classification decision. F3 and F9 push the class assignment towards C2, whereas F6 does the opposite, decreasing the likelihood of C2. Similar to F6, F1, and F4 negatively impact the C2 classification, whereas F2, F5, and F8 positively push the decision towards the C2 class. Features F7, and F10 all have little impact on the final decision, with F10 having the least impact.

{'C2': '69.02%', 'C1': '30.98%'}

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

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

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

On-time

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

RandomForestClassifier

C1

Advertisement Prediction

The classifier trained on this prediction problem assigns a label to a given case based on the information supplied. The class assigned by the classifier to the case under consideration is C1. The probability that C2 is the correct label is around 25.28%; therefore, it is less likely to be the true label. The above classification decision is mainly based on the influence of the features F1, F6, F5, F3, F4, F7, and F2. Of the above stated features, F3 and F6 are the ones shown to have a negative impact, decreasing the odds of C1 being the accurate label for the given case and encouraging the classifier to select C2 instead. Finally, it can be concluded that there is a moderately high level of confidence in the assigned label, which can be attributed to the strong positive contribution of F1 combined with other positive features such as F5 and F4.

{'C1': '74.72%', 'C2': '25.28%'}

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

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

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

Skip

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

GradientBoostingClassifier

C1

Food Ordering Customer Churn Prediction

The case given is labelled as C1 by the classifier with a confidence level equal to 82.07%. Therefore, the probability of C2 being the correct label is only 17.93%. The classification above is mainly due to the contributions of features such as F22, F14, F10, and F40. F20, F7, and F31 are the next three with moderate influence. However, not all the features are considered by the classifier when determining the correct label for the given case. F3, F43, F21, and F34 are notable irrelevant features. With regards to the direction of influence of the relevant features, F22, F14, F10, and F40 are the top features with strong positive contributions favouring the assignment of label C1. The top negative features that shift the classification in a different direction are F20, F7, F32, and F27. Considering the fact that a number of the relevant features have positive attributions, it is not surprising that the classifier is quite certain that the appropriate label is C1 instead of C2.

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

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

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

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

Go Away

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

SVM_linear

C3

Mobile Price-Range Classification

According to the algorithm, there is little to no chance that the correct label for the given data instance is any of the following classes: C4, C2, and C1. It is very confident that the proper label is C3. This label assignment is largely due to the parts played by the features F14, F16, and F1. On the lower end are the input features F3, F9, F18, and F7, which are shown to be less relevant when it comes to this labelling assignment task. Finally, among the top features identified during the attribution investogation, only F5 and F12 are features with a negative influence, decreasing the odds of C3 being the appropriate label here.

{'C4': '0.00%', 'C2': '0.00%', 'C1': '0.00%', 'C3': '100.00%'}

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

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

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

r4

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

SVC

C1

Paris House Classification

The model predicts that the label for this case is C1 with a high degree of certainty of about 99.19% and the probability of the other label is only 0.81%. From the analysis, the variables with the strongest attributions to this classification decision are F9, F17, and F4. The attributions of these variables increased the response of the model in favour of labelling the case as C1. Other variables that positively supported the label decision include F15, F2, and F11. Not all the variables support the model's prediction of C1 and this is because the values of F16, F6, F8, F1, and F5 are driving the prediction towards C2. The joint attribution from these variables is weaker than that from F9, F17, and F4, so the model is biased toward predicting C1. Finally, F13, F10, F14, and F12 are the least important positive features, given that they have minimal attributions in favour of C1.

{'C1': '99.19%', 'C2': '0.81%'}

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

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

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

Basic

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

LogisticRegression

C1

Used Cars Price-Range Prediction

According to the output prediction probabilities across the two classes, the output decision for the given data is C1 with a very high confidence level. C2 has a prediction probability of about 0.00%. The variables contributing most to the abovementioned classification are F10, F5, and F3, whereas F9 and F1 are the least influential variables. The very high confidence level associated with the classification decision here could be attributed to the fact that a greater number of the input variables have attributions that increase the model's response towards label C1. F6, F8, and F9 are the variables with negative contributions that attempt to push the model to label this case as C2. To put it in a nutshell, the joint contribution of the negative variables is very low unlike that of the positive variables, hence the model's certainty in the decision here.

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

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

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

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

Low

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

SVC

C2

Tic-Tac-Toe Strategy

With a labelling confidence level of 99.50%, the classifier predicts the label C2 in this situation. Hence, it is correct to conclude that the classifier is less certain that C1 is the proper label for the case here. The analysis indicates that five features contradict the decision above, while four features support the classifier. The features contradicting the prediction are usually referred to as negative features while those supporting it are referred to as positive features. The negative features decreasing the odds of C2 being the correct label are F2, F6, F5, F4, and F7. Conversely, the positive features increasing the odds of C2 are F8, F9, F1, and F3.

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

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

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

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

player B win

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

SVMClassifier_poly

C1

Employee Attrition

The class assigned by the model is C1 with a close to 97.67% confidence level, implying that the likelihood of C2 is only 2.33%. Based on the analysis, the most important features considered during the classification are F10, F27, F30, and F2 but among these features, F27 and F30 are the only ones with negative attributions, decreasing the likelihood of C1 being the label for the given case. Furthermore, moderately influencing the decision are F19, F21, F23, and F17. F19, F21, and F23 have positive attributions, while F17 has a negative impact, shifting the prediction in a different direction. Finally, the features with insignificant impact on the model when it comes to this case include F28, F13, F29, and F22.

{'C1': '97.67%', 'C2': '2.33%'}

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

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

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

Stay

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

KNeighborsClassifier

C1

Advertisement Prediction

With a higher degree of confidence, the model labels this given case as C1 since there is a zero chance that it is C2. The classification here can be attributed to all the features having positive contributions, decreasing the odds of C2 being the correct label. The features can be ranked based on their degree of influence from the most relevant to the least relevant as follows: F7, F6, F4, F5, F2, F1, F3. This implies that F7 is the most influential feature, while F3 is the least influential among the input features.

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

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

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

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

Watch

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

SVM_poly

C4

Mobile Price-Range Classification

According to the model, C4 has a prediction probability of 99.45 percent, C3 has a prediction probability of 0.47 percent, C2 has a prediction probability of 0.04 percent, and C1 has a prediction probability of 0.05 percent, therefore, the most likely class is C4. F2 and F17 positively influence the above-mentioned label decision in favour of C4, but F13 has the opposite effect, favouring a different label. F9 and F1 both have a similar negative impact on the C4 prediction, whereas F20 has a positive impact. In this case, F18, F19, F3, and F6 have little influence on the labelling result. All in all, the model is confident in its assignment of the C4 class as shown by the predicted probabilities across the classes.

{'C4': '99.45%', 'C3': '0.47%', 'C2': '0.04%', 'C1': '0.05%'}

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

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

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

r1

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

GradientBoostingClassifier

C2

Food Ordering Customer Churn Prediction

Per the model employed here, the prediction probability of C1 is only 17.93%, and that of C2 is equal to 82.07%. Given the information provided to the model, the most valid conclusion regarding the true label is that C2 is without a doubt the most likely one. The attributions analysis indicates that F20, F30, F11, F13, and F43 are the major drivers resulting in the prediction probabilities across the classes under consideration. At the tail end are features such as F44, F10, F41, and F36 that have very little influence on the decision made with respect to the given case. Among the influential features, only F20, F30, F13, F16, F21, F46, F14, F24, and F42 have positive contributions in support of labelling the given case as C2. On the other hand, the negative features such as F11, F43, F26, F45, F3, F32, and F17, suggest C1 could likely be the true label in this case. Overall, the marginal doubt in the correctness of assigning C2 to the case under consideration is attributed to the negative features driving the model's decision in the direction of C1 away from C2. But the higher influence of positive features such as F20 and F30 ensures that C2 is assigned as the most probable label.

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

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

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

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

Return

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

RandomForestClassifier

C2

Personal Loan Modelling

The model is about 90.0% certain or sure that the correct label based on the input features of the given case is C2. The features with the most significant influence on the decision are F3, F2, F6, and F1. The influence of the features can be categorised as positive or negative traits depending on the direction of the effect on the model. Positive features increase the likelihood of the most likely class (i.e., C2), whereas negative features reduce the model's responsiveness to the assigned label, favouring the less likely class (i.e., C1). From the attribution analysis, F8, F7, and F5 are the negative features here. Overall, the negative features are shown to have moderate to low influence compared to the positive features, hence explaining why the model is very confident about the assigned label C2.

{'C1': '10.00%', 'C2': '90.00%'}

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

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

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

Accept

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

LogisticRegression

C1

Tic-Tac-Toe Strategy

With an 81.01% chance of being correct, C1 is the most likely label, consequently, the C2 class's prediction probability is only 18.99%. The algorithm or classifier got the above prediction mostly due to the influence of features like F2, F7, F8, and F6. F1, which is found to have very little impact with regard to the label choice here, is the least relevant feature for the algorithm. F7, F9, F8, and F6 have a positive direction of influence, pushing the algorithm higher towards the C1 label. Negative features like F2, F3, and F4 favour choosing or labelling the case as C2.

{'C2': '18.99%', 'C1': '81.01%'}

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

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

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

player B win

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

SVC

C1

Student Job Placement

The model makes classification decisions based on the information provided to it and for the case here, the prediction probabilities across the two class labels, C2 and C1, are 49.32% and 50.68%, respectively. Based on these prediction probabilities, the label assigned is C1, since it has the highest likelihood, however, the model is not very certain about the correctness of the assigned label since its probability is marginally higher than the average. The uncertainty in the classification here can be blamed on the fact that only F10, F2, F12, F6, and F8 have positive attributions, shifting the decision higher towards C1. On the other hand, features F7, F3, F1, F5, F9, F11, and F4 have negative contributions that decrease the prediction likelihood of C1 while increasing that of C2. To cut a long story short, the most positive features are F10 and F2, whereas the most negative ones are F7 and F3. Finally, F9, F8, and F11 are not as important as all the previously mentioned features hence received little attention from the model.

{'C2': '49.32%', 'C1': '50.68%'}

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

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

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

Placed

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

SVMClassifier_liner

C1

Employee Attrition

The most likely label for the given case is C1 since the predicted probability of C2 is only 34.27% and this means that the likelihood of C1 is 65.73%. The most relevant features that led to the C1 classification verdict are F22, F6, F17, F2, and F21. However, some of the features are deemed irrelevant to the above verdict and these include F1, F16, F10, and F5. Among the relevant features with some degree of impact, seven are shown to drive the model's class assignment towards the C2, while the remaining support the C1 prediction. Notable negative features swinging the prediction towards C2 are F22, F6, and F17, while the notable positive features are F2 and F21. The small uncertainty associated with the prediction decision for the given case could be attributed to the fact that all the three most important features are negative features whose values contradict assigning the label C1.

{'C1': '65.73%', 'C2': '34.27%'}

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

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

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

Stay

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

RandomForestClassifier

C1

Printer Sales

Per the classifier for the given data, the most plausible label is C1. F17, F8, F6, and F3 are the main features pushing for the above-mentioned outcome. F4, F19, F9, F21, F22, and F11, on the other hand, have little contribution to the classifier employed here. F2, F10, F26, and F14 have a moderate contribution to the assignment of C1. The classifier's confidence in the label decision above can be attributed to larger positive attributions of F10, F2, F6, and F8 compared to the negative attributions of F26, F5, F17, F1, F3, and F24.

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

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

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

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

More

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

SGDClassifier

C2

Job Change of Data Scientists

The least probable class, according to the classification algorithm, is C1, with a prediction probability of 25.12%, therefore, we can conclude that the algorithm is quite confident that the correct label for this data is C2. Analysing the attributions revealed that F12, F7, F1, and F3 are the most relevant features, whereas F2, F6, and F4 are the least relevant features. Increasing the algorithm's response in favour of C2 are the positive features F12, F1, F3, F6, F2, and F10. On the contrary, all the other features, F7, F5, F9, F11, F8, and F4, drive the algorithm towards labelling the given data as C1, hence they are considered negative features. Furthermore, the negative influence on the algorithm is the reason why the confidence level in the C2 is reduced to 74.88%.

{'C2': '74.88%', 'C1': '25.12%'}

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

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

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

Stay

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

SVM_poly

C1

Mobile Price-Range Classification

According to the classification algorithm, neither C2 nor C3 nor C4 is the correct label for the given case. It is 100.0% certain that C1 is the right label. The higher degree of certainty in the above prediction can be attributed to the positive contributions of F19, F20, and F4. The other positive features include F14, F8, F17, and F9, however, unlike F19, F20, and F4, these features have a moderately low impact on the algorithm's decision. The remaining positive features, F6, F18, F7, and F3, are among the least influential input features considered by the algorithm. There are other features such as F2, F1, F10, and F15 whose contributions only serve to decrease the odds of C1 being the correct label for the given case. Regarding the high confidence of the algorithm with respect to this classification, one can conclude that the negative features have little influence on the algorithm's label decision here.

{'C2': '0.00%', 'C3': '0.00%', 'C4': '0.00%', 'C1': '100.00%'}

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

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

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

r4

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

DNN

C1

Ethereum Fraud Detection

The prediction likelihoods across the two classes are 15.35% for class C2 and 84.65% for C1, it can be concluded that C1 is the most probable class label for the given data instance. According to the attribution analysis conducted, the different input variables have varying degrees of influence on the model's decision here. The most influential set of variables is F1, F37, F19, F34, F24, F36, and F11, while the variables with the least influence include F9, F12, F27, F17, F14, and F38. The following or subsequent analysis performed to understand the direction of contribution of of the features will focus on the most influential ones controlling the label selection here. Among the top influential features, F1, F37, F19, F34, and F11, only F1 and F37 have negative contributions, decreasing the probability that C1 is the correct label, and they strongly support labelling the case as C2 instead. Pushing the classification decision in favour of C1 are the positive variables such as F19, F34, and F11. The contributions of the remaining variables, including F24, F36, and F31, have moderate to low influence. All in all, the marginal uncertainty in the decision here is mainly due to the negative influences of F1, F37, F29, and F3, but the positive contributions of F19, F34, F31, F24, F36, and F11 drive the decision higher towards C1.

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

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

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

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

Fraud

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

KNeighborsClassifier

C2

Credit Risk Classification

According to the machine learning model, it is more likely that the case's label is C2, with a certainty of 100.0%, and this prediction decision is mainly based on the effects of the following features: F10, F4, F1, F8, and F9 on the model. Apart from F9 and F8, all the other variables mentioned above have a strong positive influence, improving the odds of the prediction class, C2. Together with F9 and F8, the values of variables F2 and F11 indicate that C1 could be the correct label instead. Unlike the top positive variables, F10, F4, and F1, each of these negative variables has a moderate contribution to the final decision. The features F3, F5, F6, and F7 are shown to have made minor contributions to the model's decision in this case. In summary, with only the positive contributions from F10, F4, F1, F6, and F3, the model is very certain of the classification output as indicated by the predicted probabilities across C2 and C1.

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

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

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

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

Low

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

SVMClassifier_poly

C1

Employee Attrition

The classification findings by the model for the case here are as follows: there is a 97.67% chance that C1 is the correct label hence only a marginally low chance of 2.33% that C1 is not the correct label but C2 is. From the above findings, it is valid to conclude that the right class for the given case is C1, and the model is very certain of this decision. The features with the most control and influence on the classification above are F16, F23, F17, F19, and F28 but the influence of the remaining features is either moderate or low or negligible. Some of the features with moderate impact include F4, F13, F9, and F20. Those with low influence are F25, F18, F26, F5, and F24. Finally, those with negligible impact are F30, F21, F12, F27, F11, F3, F8, F29, F7, and F15 since their values are shown to have no impact on the classification made by the model here. The top positive features increasing the prediction likelihood of class C1 are F16, F28, and F14. Conversely, the negative features decreasing the odds in favour of C2 are primarily F23, F9, and F17.

{'C1': '97.67%', 'C2': '2.33%'}

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

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

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

Stay

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

LogisticRegression

C1

Flight Price-Range Classification

The model is very confident that C1 is the most probable class for the given case, with a probability of 90.48% which means that the other labels are very unlikely. F11 and F9 are the most important variables with respect to this classification verdict while all other variables are shown to have a medium or low impact. Fortunately, the top variables, F11 and F9, have the same direction of influence, increasing the likelihood of C1. Furthermore, while F4 and F8 push the model to predict C1, those pushing for the assignment of a different label are F12, F2, and F3. Finally, many features have a fairly small impact on the final prediction made by the model here, but F6, F2, and F7 have the least impact.

{'C1': '90.48%', 'C3': '9.51%', 'C2': '0.01%'}

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

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

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

Low

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

SVC

C1

Water Quality Classification

Despite the reasonably high confidence in the assigned label, the prediction probabilities across the two classes indicate that C2 might be the correct label. F3, F7, F6, and F4 are the factors whose major contributions resulted in the labelling choice mentioned above. According to the analysis, the top two factors, F3 and F7, have a negative influence, leading the classifier to classify the data as C2 rather than C1. F9 is the only other negative variable with a moderate effect when compared to the other two negative variables. Nevertheless, there are several factors, F6, F4, F1, F2, F5, and F8, that favourably support and encourage the classifier to assign C1. All in all, the degree of uncertainty in this classification instance might be explained by just looking at the negative factors' rather strong pull on the classifier towards C2.

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

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

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

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

Portable

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

MLPClassifier

C1

Ethereum Fraud Detection

C2 has a probability estimate of only 6.80%, while that of C1 is 93.20%; consequently, the most likely class for the given case is C1. The important or relevant features considered by the classifier are F12, F35, F5, F8, F10, F9, F19, F1, F26, F11, F27, F30, F15, F38, F37, F7, F18, F25, F20, and F33. Not all input features are relevant when determining the appropriate label and these irrelevant features include F32, F3, and F24. Furthermore, F12 and F35 have a strong positive effect, increasing the odds in favour of C1. In contrast, the F5, F10, and F8 are the negative features, lowering the odds of C1. Comparing the attributions of F12, F9, and F35 features to those of the negative features mentioned above, it is not surprising that the classifier is convinced that C1 is the most likely label here.

{'C2': '6.80%', 'C1': '93.20%'}

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

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

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

Fraud

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

BernoulliNB

C2

German Credit Evaluation

The model is not 100% convinced that the correct label for the data under consideration is C2 since there is a 26.27% chance that labelling the data as C1 is correct. All the input variables are shown to have some degree of influence on the classification decision, with the most influential variables being F4, F3, and F2, whereas F7 and F6 are the least influential. The impact of F5, F9, F1, and F8 can be considered moderate compared to the F4, F3, and F2. The uncertainty surrounding the above classification can be blamed on the fact that the majority of input variables have values suggesting that C1 could be the appropriate label. The negative features that decrease the prediction likelihood of C2 are F4, F2, F1, and F8. However, given that the prediction probability is about 73.73%, it can be said that the influence of positive features, F3, F5, F9, and F7, is enough to swing the model's verdict in favour of C2.

{'C2': '73.73%', 'C1': '26.27%'}

{'F4': 'Saving accounts', 'F3': 'Sex', 'F2': 'Housing', 'F5': 'Purpose', 'F9': 'Checking account', 'F1': 'Job', 'F8': 'Duration', 'F7': 'Age', 'F6': 'Credit amount'}

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

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

Good Credit

{'C2': 'Good Credit', 'C1': 'Bad Credit'}

SVMClassifier_poly

C2

Employee Attrition

The model predicted class C2 with an 81.98% prediction likelihood. F13 had the largest impact, followed by F11, F1, F9, F19, F16, F7, F24, F3, F6, F25, F18, F17, F5, F23, F21, F14, F22, F26, and finally, F2, which had the smallest non-zero impact. F13, the feature with the largest impact, contributed against the direction of the prediction, whereas F11, F1, F9, and F19 all contributed positively towards the prediction. Other features that had a negative influence on the prediction included F7 and F24, whereas F16 had a positive influence on the prediction. F8, F10, F12, and F20 are shown to have close to zero attribution in the model's prediction verdict in the given case.

{'C2': '81.98%', 'C1': '18.02%'}

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

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

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

Stay

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

SVC

C1

German Credit Evaluation

This case's label has a 70.83 percent chance of being C1 and per the predicted likelihoods across the alternative labels, C2 has a 29.71 percent chance of being the correct label, however, the model is certain that C3 is not the true label. The most important variables are F7, F2, F9, and F4, whereas the remaining influential variables are listed in order of the magnitude of their contributions: F3, F8, F1, F5, and F6. Three of the nine variables have values that push towards the prediction of label C2 while the other attributes are referred to as positive since their values inspire the prediction of class C1. F7, F2, and F9 are the three attributes that have a negative influence on the prediction judgement, pushing it away from C1 towards the label C2. Finally, it is essential to highlight that the cumulative effect of positive attributes is greater than that of negative attributes, F9, F2, and F7.

{'C1': '70.83%', 'C2': '29.17%', 'C3': '0.0%'}

{'F4': 'Checking account', 'F7': 'Duration', 'F2': 'Housing', 'F9': 'Saving accounts', 'F3': 'Sex', 'F8': 'Age', 'F1': 'Purpose', 'F5': 'Job', 'F6': 'Credit amount'}

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

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

Good Credit

{'C1': 'Good Credit', 'C2': 'Bad Credit', 'C3': 'Other'}

SVC

C1

Vehicle Insurance Claims

First of all, the classification decision is solely based on the information or data supplied to the prediction model. According to the model, there is a 61.61% chance that C1 is the true label, and a 38.39% chance that C2 is the true label. Since the predicted probability of C1 is higher than that of C2, it is valid to conclude that C1 is most likely the true label. The main feature responsible for this classification is F10, with a very strong positive influence, driving the model's decision higher towards C1. The next set of relevant features are F30, F25, F1, F18, F14, F19, F31, and F3. Among all the features mentioned above, F30, F1, F18, F19, and F31 have negative contributions that are responsible for the decrease in the probability that C1 is the true label. This implies that the contributions of F25, F14, and F3 combined with that of F10 explain why the model is moderately certain that C1 is the true label.

{'C1': '61.61%', 'C2': '38.39%'}

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

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

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

Fraud

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

GradientBoostingClassifier

C1

Paris House Classification

Because the prediction probability of C2 is barely 0.70 percent, the classifier outputs the label C1 with near 100 percent confidence based on the values of the input attributes. The effects of F3, F6, and F8 on the aforementioned classification decision are significant. The values of these features are given greater emphasis by the classifier than the others. F8 is has a negative impact among these top features, pushing the prediction judgement towards the least likely class, C2 whereas on the other hand, F3 and F6 are referred to as positive features since they improve the likelihood of the C1 label rather than the C2 label. Finally, unlike the others, the values of F14, F17, F10, and F12 have only a little influence on the label selection made here.

{'C1': '99.30%', 'C2': '0.70%'}

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

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

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

Basic

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

SGDClassifier

C3

Flight Price-Range Classification

The classification algorithm arrived at the prediction output based on the variables or information supplied about the case under consideration. The prediction probabilities across the three-class labels, C2, C3, and C1, respectively, are 28.17%, 50.21%, and 21.62%, making C3 the label assigned by the algorithm, judged based on the prediction probabilities. The attributions analysis suggests that F2, F9, F8, and F7 are the positive features that increase the algorithm's prediction response in favour of C3. On the other hand, F4, F6, F11, F3, F5, F12, F10, and F1 have negative contributions in support of labelling the case as either C2 or C1. Overall, judging by the degree of contributions of the positive features, it is not surprising that the algorithm is moderately certain that neither C2 nor C1 is the most probable label for the case under consideration here.

{'C2': '28.17%', 'C3': '50.21%', 'C1': '21.62%'}

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

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

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

Moderate

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

RandomForestClassifier

C1

Paris House Classification

Judging based on the information provided on the case under consideration, the model outputs that the prediction probability of C2 is only 0.48%, indicating that with about 99.52% certainty, the true label here is C1 and in simple terms, the model is very confident that the true label for the case under consideration is C1. The higher degree of certainty in the above classification can be attributed solely to the positive contributions of influential features F16, F12, and F8. Analysis indicates that all the remaining features such as F1, F6, F3, F7, and F5 have moderate to low contributions towards the prediction conclusions above, whereas F14, F2, F17, and F15 are the least relevant features here. The very marginal decrease in the C1's prediction likelihood could be attributed to the influence of negative features F6, F7, F10, F15, and F2 since their contributions support labelling the case as C2 instead. Moderate positive features further driving the model to label this case as C1 are F1, F3, F13, and F5.

{'C1': '99.52%', 'C2': '0.48%'}

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

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

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

Basic

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

GradientBoostingClassifier

C2

Basketball Players Career Length Prediction

Judging based on the values of the variables passed to the model with respect to the case under consideration, the output labelling decision is as follows: there is about an 83.98% chance that C2 is the correct label, whereas the likelihood of C1 is only 16.02%, hence the label choice with a higher confidence level is C2. The top-variables influencing this decision are F17, F1, F19, and F11, while the least important variables are F6, F8, and F10. According to the variable contributions analysis performed, only the input variables F18, F3, F13, and F2 exhibit negative attributions, pushing the prediction decision towards the alternative label, C1. The other variables positively support the C2 prediction, shifting the verdict strongly away from the C1 class. In conclusion, positive variables such as F17, F1, F19, F11, F5, and F14 have a higher joint contribution compared to the negative features, which can explain why the model is certain that C2 is the most probable label.

{'C2': '83.98%', 'C1': '16.02%'}

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

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

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

More than 5

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

RandomForestClassifier

C4

Mobile Price-Range Classification

The model predicts the class label C4 for the given test instance with a likelihood of about 69.23%. However, there is about a 30.77% chance that the true class label is C1, while the others, C2 and C3, have a 0.0% likelihood. The top features contributing to this prediction decision are F14, F9, F11, and F15, whereas the least important are F8, F4, and F7. Among the top features, while F14 and F9 have values that shift the prediction decision towards the C4 class label, the values of F11 and F15 suggest that the true label could likely be C1. For the features with moderate influence on the decision, F19, F20, F5, and F3 have negative contributions, further decreasing the confidence level in the C4 assignment. On the other hand, the moderate positive influences of F17, F18, F13, F1, and F10 drive the decision further towards the C4 label. Considering the attributions of the input features, it is surprising that the confidence level is just 69.23% since the top feature, F14, has the highest contribution among all the input features. Finally, the values of F2, F4, and F7, though shown to be less important when deciding the correct label for the given case, have positive contributions to the prediction with respect to the given case.

{'C2': '0.00%', 'C4': '69.23%', 'C1': '30.77%', 'C3': '0.00%'}

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

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

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

r2

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

KNeighborsClassifier

C2

Water Quality Classification

The given case is likely C2 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, F6 and F8 have a high degree of impact. F9, F7, F5, F1, and F2 have a moderate degree of impact while on the contrary F3 and F4 have little impact. Examining further, the values of F6, F8, F9, and F7 all have a positive influence on the classifier supporting the label assignment decision for the given test case. F5 and F2 are also positively supporting features, whereas F1 has a negative influence on the final classification. Finally, F3 and F4 both have very little contributions, though F4 has significantly less than even F3.

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

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

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

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

Not Portable

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

RandomForestClassifier

C2

Mobile Price-Range Classification

The label for this example is estimated to be C2 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 F16, F13, F18, F3, and F17. F16 had the greatest influence, followed by F18, F13, F17, and F3. The positive variables F16, F13, F10, and F7 outnumber the negative variables F18, F17, F3, and F14. 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 C2 assigned. Given that the chance of C2's being accurate is 73.08 percent, the model appears to be relatively confident in its final judgement for the data instance under review.

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

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

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

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

r1

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

BernoulliNB

C1

Personal Loan Modelling

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

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

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

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

{'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%. F2 and F3 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: F10 and F12 have a negative effect, while F15 and F14 have a positive effect on the prediction of C2. Similar to F10 and F12, the features F13 and F4 also negatively affected the prediction decision. Finally, the values of F16, F1, F11, and F6 are the least important to the model decision for this case.

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

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

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

{'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%. F17 is by far the most influential feature whereas F5, F7, and F18 have been recognised as having the biggest effect on prediction output here after F17. The combination of F17, F5, F7, F18, and F4 features has resulted in the classification choice being altered from C2 to C1. While F10, F1, and F19 all have a minor influence on the classification, F10 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, F16, F12, F8, and F2 having a marginal effect.

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

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

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

{'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 F4, F3, and F7, however, F2, F10, and F9 are shown to be the least important variables. Regarding the direction of influence of the variables, F4, F7, F6, F2, and F10 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 F3, F8, F5, F1, and F9. Owing to the fact that the most influential variables, F4 and F7, 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%'}

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

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

{'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 F5, F3, F4, and F1, followed by F8, F2, F9, F7, and finally F6. Based on the inspections performed to understand the direction of influence of the input features, it can be concluded that F5 has the strongest positive contribution, while F4 has the strongest negative contribution and conversely, all the remaining features have moderate contributions. The other positive features are F3, F8, F2, and F7, whereas the remaining negatives are F1, F9, 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%'}

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

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

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

Not Portable

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

RandomForestClassifier

C3

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, C3, C2, and C1, the model shows without a doubt that neither C2 nor C1 is the true label, given that the probability of C3 being the true label is 100.0%. F9, F1, and F11 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 F12, F2, and F10 are the least influential variables since they receive little emphasis from the model when making the labelling decision here. In between F9, F1, and F11, and F12, F2, F5, and F10, are the variables such as F7, F4, F3, and F8 with moderate influence on the classification decision here. Among the variables passed to the model, only F7, F8, and F12 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 C3, it is reasonable to deduce that the positive variables, such as F9, F1, F11, F4, F6, and F3, significantly influence the model's judgement towards C3.

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

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

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

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

Low

{'C3': 'Low', 'C2': 'Moderate', 'C1': '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: F13, F8, F5, F18, F11, F3, and F7. The top negative variables decreasing the likelihood of C2 are F4 and F6 supported by other negative variables, F1, F14, and F2, that further shift the verdict towards C1.

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

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

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

{'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. F19, F6, and F8 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 F2, F24, F18, F14, F13, and F12 when giving a label to this case since their relative degrees of impact are extremely near to zero. F1, F16, F23, F21, F7, and F4 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%'}

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

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

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

Less

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

RandomForestClassifier

C1

Credit Risk Classification

According to the ML model, C1 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 C2 as the correct label is only 7.0%. For the case under study, analysis indicates that F7, F11, F9, and F2 are essentially the negative set of features that push the forecast higher towards C2 instead of C1, while F5, F6, F4, and F10 increase the odds of the prediction being equal to C1. In general, the most relevant feature is F5, while F8 and F3 are the least relevant features, with marginal influence on the above classification verdict. In summary, given the very strong positive influence of F5 together with the moderate influence of the other positives, F6, F10, and F4, it is not strange that the model chose to label the case as C1 instead of C2.

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

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

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

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

Low

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

MLPClassifier

C2

Annual Income Earnings

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

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

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

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

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

Above 50K

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

KNNClassifier

C2

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 C2 and C1 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 F5 and F3, with F1 and F2 ranked as the least contributing factors. The values of F6 and F4 suggest that perhaps the true label could be C1 since they are the negative features. However, considering the confidence in C2, 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 F5, F3, F1, and F2.

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

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

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

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

Unacceptable

{'C2': 'Unacceptable', 'C1': '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 F18, F12, F1, and F20, while on the other hand, the least relevant features with little contributions to the decision based on the analysis are F15, F19, F2, and F4. The top positive features Increasing the likelihood of the prediction being made are F18, F12, and F20. Pushing the prediction towards the alternative class C2, the top negative features are F1, F13, and F3. F11, F7, F10, F6, and F17 are some of the features that have a moderate impact on the classification decision in this case.

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

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

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

{'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 F6, F9, F13, F21, and F33. The most negative feature is F6, 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 F6 and F13 is somewhat counterbalanced by the values of the features F9, F21, and F33. Other attributes that shift the decision in favour of C1 are F11 and F15. F31 shifts the decision further in the direction of C2 and in addition, F3 supports the model's prediction while the values of F29 and F32 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 F25, F22, F12, and F2.

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

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

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

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

Not Fraud

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

RandomForestClassifier

C1

Ethereum Fraud Detection

According to the classification algorithm, the best label for the given case is C1, because there is little to no chance that C2 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: F24, F20, F13, F34, F6, F38, F31, F33, F26, F25, F23, F28, F9, F18, F35, F12, F15, F1, F8, F16. F3, F27, and F30, on the other hand, are unimportant features since they have almost no influence. Among the most influential features F24, F20, F13, F34, and F6, F13 is considered the most negative, dragging the verdict in a different direction, while the others have positive contributions, increasing the possibility that C1 is correct in this case. F31 is recognised as a positive feature with modest effect, whereas F38 and F33 are identified as negative features. Given that the majority of the top five attributes have positive contributions, boosting the likelihood that C1 is the correct label, it is not unexpected that the algorithm is quite confident in the assigned label's accuracy.

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

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

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

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

Fraud

{'C2': 'Not Fraud', 'C1': '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: F8, F7, F4, F11, F6, F10, F3, F5, F2, F1, F12, F15, F13, F9, and F14. Among the top features, F8 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 F7, F4, and F11. Similar to F8, the features F3, F9, and F1 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%'}

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

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

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

dissatisfied

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

RandomForestClassifier

C2

Used Cars Price-Range Prediction

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

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

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

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

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

High

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

SVC

C2

Food Ordering Customer Churn Prediction

The model labels the case as C2 with fairly high confidence equal to 89.73%, whereas the likelihood of C1 is only 10.27%. Analysis shows that only 20 of the 46 input variables contribute to the prediction assertion above. The prediction judgement C2 is mainly based on the variables F27, F11, F24, and F2. F46, F28, F7, F32, F43, and F6 also contribute to the decision, however, their degree of influence is only moderate. According to the direction of influence analysis, F27, F2, F43, and F32 positively support the decision of the model to assign the label C2. However, F11, F28, F6, F24, F46, and F7 reduce the likelihood or chance that C2 is the true label for this particular test instance. The main variables with less influence on the above classification decision are F38, F36, F19, and F31.

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

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

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

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

Return

{'C2': 'Return', 'C1': '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 F7, F4, F14, and F2. On the other hand, F1 and F6 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 F14, F10, F12, F13, F8, and F6 and countering their influence are the top positive features are F7, F4, F5, and F2.

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

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

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

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

Above 50K

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

SVM_linear

C2

Wine Quality Prediction

The likelihood of C2 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 C1 could be the label and the classification assertion above is influenced mainly by the variables F1, F7, F11, and F5. On the contrary, F9, F6, and F4 are deemed less important when deciding the correct label for this given case. Decreasing the likelihood of the predicted label , C2, are the variables F5, F3, F6, and F4, therefore, these negative variables support the alternative class C1. However, the collective or joint attribution of the top positive variables, F7, F1, and F11 is strong enough to tilt the classification in favour of C2.

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

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

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

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

high quality

{'C1': 'low_quality', 'C2': '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: F6, F3, F2, F9, F5, F4, and F1. According to the analysis, the features F6, F3, and F2 have a very strong positive influence, swinging the prediction decision towards C1. In contrast, the value of F9 also suggests the decision should be the alternative class, C2. Similar to F9, the values of F7, F5, and F4 indicate the label could be C2. However, the influence of these features is very small compared to F6, F3, F2, and F9. Finally, the attributes with a moderately low influence on the final prediction decision for this case include F1, F10, F11, and F8. The values of F1 and F8 have a negative attribution, while F10 and F11 have positive attributions.

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

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

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

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

Low

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

BernoulliNB

C2

Job Change of Data Scientists

The classification algorithm is pretty confident that the correct label for the data under consideration is C2, wowever, it is noteworthy to consider that C1 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 F1, F4, F9, and F12. However, the analysis shows that the values of F8, F6, F7, and F3 are less relevant when classifying the data. Only the features F9, F2, F8, F6, F7, and F3 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 C1.

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

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

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

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

Leave

{'C1': 'Stay', 'C2': '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 F1, F25, and F14 were the most important features driving the model to arrive at the labelling assignment of C2. F16 and F4 have nearly identical positive attributions, while F19 and F7 has negative impacts, swinging the prediction towards a different label. However, the joint positive impact of F16, F1, F14, and F4 stands out over the impact of F25, F11, F7, and F19, 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 F10, F8, F26, and F20 hence they are practically irrelevant when it comes to labelling this case.

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

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

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

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

More

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