Harsh2396/gsk-copay-fraud-detection

GSK Copay Card Fraud Detection System — v4 Group-Aware

Product Focus: Trelegy Ellipta (configurable for Nucala / any GSK product)
Methodology: Hybrid Rules + Isolation Forest + SHAP Explainability + Hierarchical Summaries + Group-Aware Benefit Validation + EDA
Architecture: Drug-agnostic, ground-truth-optional, vendor-format-agnostic, production-ready
Analytical Levels: Transaction → Patient → HCP → Pharmacy
Group-Aware Validation: Group 8141 (Legacy) vs Group 8200 / 2025 benefit designs

---

Overview

This system detects fraudulent and suspicious copay card claims in GSK pharmaceutical transaction data using a 4-level hierarchical analytical framework:

Level	What It Detects	Key Features
Transaction	Per-claim anomalies	Gap between fills, quantity, days supply, benefit amount, OOP cost, NDC switch
Patient	Behavioral patterns	One-and-done patients, active duration, avg gap between fills, short/long gap %
HCP	Prescriber-driven fraud	Suspicious specialty, one-and-done %, patient concentration, avg benefit per patient
Pharmacy	Pharmacy-centric rings	Active/closed flag, HCP concentration, one-and-done %, avg benefit, fraud risk score

Under the hood, the system combines:

• 35 hard-coded business rules (23 original + 12 v4 group-aware rules)
• Isolation Forest unsupervised anomaly detection trained on rule-clean data
• SHAP explainability for every flagged claim
• Hierarchical summary exports for investigative lens views
• Exploratory Data Analysis (EDA) module for pre-modeling data profiling

The pipeline supports any vendor format — ELAAD, APLD, IQVIA, CMS DMR, generic CSV. It auto-discovers column names via synonym mapping, handles missing columns gracefully, and produces a schema report showing what it found and what it missed.

---

Project Structure

gsk_copay_fraud/
├── config.py                          # Product config + column synonym mappings + FEATURE_DEPENDENCIES
├── data_ingestion.py                  # Schema discovery + vendor-agnostic ingestion
├── feature_engineering_v2.py          # 60+ features with graceful degradation
├── fraud_detection_pipeline_v3.py     # Full pipeline (35 rules + IF + SHAP + hierarchical summaries)
├── eda.py                             # NEW: Exploratory Data Analysis module
├── run_all_v3.py                      # CLI runner (now supports --run-eda)
├── generate_elaad_test_data.py        # Synthetic ELAAD-style test data generator with embedded fraud
├── requirements.txt                   # Dependencies
├── README.md                          # This file
├── data/                              # Place raw GSK data here
│   └── elaad_test_trelegy.csv
└── results/                           # All outputs generated here
    ├── eda/                           # NEW: EDA outputs (JSON + PNG)
    │   ├── 00_data_quality.json
    │   ├── 01_temporal_analysis.png
    │   ├── 02_financial_analysis.png
    │   ├── 03_categorical_analysis.png
    │   ├── 04_geographic_analysis.png
    │   ├── 05_correlation_analysis.png
    │   ├── 06_hierarchical_analysis.png
    │   └── 07_feature_engineered_eda.png
    ├── investigation_queue_top500.csv
    ├── scored_claims_full.csv
    ├── transaction_level_summary.csv
    ├── hcp_level_summary.csv
    ├── pharmacy_level_summary.csv
    ├── patient_level_summary.csv
    ├── feature_importance_shap.csv
    ├── metrics.json
    ├── schema_report.json
    ├── 01_evaluation_metrics.png
    ├── 02_shap_summary.png
    ├── 03_shap_bar_importance.png
    ├── 04_rule_breakdown.png
    ├── 05_risk_tier_distribution.png
    ├── 06_hcp_summary.png
    ├── 07_pharmacy_summary.png
    └── 08_patient_summary.png

---

Quick Start

1. Install Dependencies

pip install -r requirements.txt

2. Run EDA (Pre-Modeling Data Profiling)

# Standalone EDA
python eda.py --data-path data/elaad_test_trelegy.csv --file-type csv --vendor-format elaad_apld

# EDA + full fraud detection pipeline
python run_all_v3.py --data-path data/elaad_test_trelegy.csv --run-eda

# EDA only (skip feature-engineered EDA for speed)
python run_all_v3.py --data-path data/elaad_test_trelegy.csv --run-eda --skip-post-engineering-eda

3. Run Fraud Detection Pipeline

# Auto-detect everything (recommended)
python run_all_v3.py \
  --data-path data/vendor_file.txt.gz

# CSV file
python run_all_v3.py \
  --data-path data/gsk_copay_transactions.csv \
  --file-type csv

# Gzipped TXT (common GSK format)
python run_all_v3.py \
  --data-path data/GSK_COPAY_TRANSACTION_DAILY_20250725.TXT.GZ \
  --file-type txt.gz

# Force a specific vendor format (skips auto-discovery)
python run_all_v3.py \
  --data-path data/vendor_file.csv \
  --vendor-format gsk_iqvia \
  --contamination 0.03

# Adjust anomaly rate for unusual datasets
python run_all_v3.py \
  --data-path data/vendor_file.csv \
  --contamination 0.05

4. Generate & Test on Synthetic ELAAD Data

# Generate test data
python generate_elaad_test_data.py
# Creates data/elaad_test_trelegy.csv with embedded fraud patterns

# Run pipeline on synthetic data (high contamination needed because ~40% fraud)
python run_all_v3.py \
  --data-path data/elaad_test_trelegy.csv \
  --file-type csv \
  --contamination 0.40

---

EDA Module (eda.py)

The EDA module runs before the fraud detection pipeline to provide data quality assessment, trend analysis, and distribution profiling. It is critical for:

• Identifying data quality issues before modeling
• Understanding seasonal/temporal fraud patterns
• Validating financial distributions against expectations
• Detecting categorical imbalances (scenarios, specialties, groups)
• Finding geographic anomalies
• Establishing hierarchical relationship baselines

EDA Coverage

Analysis	Output	Description
Data Quality	00_data_quality.json	Missing values, cardinality, numeric ranges, outlier counts, cross-state summary
Temporal	01_temporal_analysis.png + .json	Monthly claim volume, day-of-week, quarterly, monthly benefit trend
Financial	02_financial_analysis.png	Benefit, copay, OOP, usual customary distributions + boxplots; benefit by scenario/group
Categorical	03_categorical_analysis.png + .json	Top 15 distributions: specialty, scenario, NDC, pharmacy type, group, insurance
Geographic	04_geographic_analysis.png + .json	Top 20 states, patient vs pharmacy state heatmap, cross-state claim count
Correlation	05_correlation_analysis.png + .json	Feature correlation matrix (lower triangle), top 30 correlated pairs
Hierarchical	06_hierarchical_analysis.png + .json	Claims per patient, patients per HCP, HCPs per pharmacy, pharmacies per patient
Feature-Engineered	07_feature_engineered_eda.png	Cap utilization ratio, early refill flags, excess payment, scenario not covered

Running EDA

# Standalone
python eda.py --data-path data/elaad_test_trelegy.csv --file-type csv --vendor-format elaad_apld

# Options
python eda.py \
  --data-path data/elaad_test_trelegy.csv \
  --file-type csv \
  --vendor-format elaad_apld \
  --results-dir results \
  --skip-post-engineering  # Skip feature-engineered EDA for speed

---

Vendor Format Handling

The Problem

Vendor files rarely match the idealised spec. A column named IQVIA_PATIENT_ID in the spec might appear as:

• PATIENT_ID (ELAAD format)
• MEMBER_ID (APLD format)
• PATIENTID (no underscore)
• PAT ID (space instead of underscore)
• Patient ID (mixed case)
• PATIENT_KEY (different suffix)

The Solution: Schema Discovery

The pipeline uses COLUMN_SYNONYMS in config.py — a dictionary where each internal column name maps to a list of possible raw names (20+ synonyms per column). When a file is loaded:

1. Scan header → collect all raw column names
2. Normalize → uppercase, strip whitespace, replace underscores/spaces with single space
3. Match → for each internal column, try synonyms in order of preference
4. Report → log what was mapped and what was missing
5. Continue → pipeline runs with whatever columns are available

Example Schema Report

2025-04-05 12:00:00 [INFO] Vendor format: detected=generic_csv, requested=auto
2025-04-05 12:00:00 [INFO] Schema discovery: mapped 38 / 50 internal columns
2025-04-05 12:00:00 [INFO] [pharmacy] 6/7 present. Missing: ['pharmacy_subcategory']
2025-04-05 12:00:00 [INFO] [reject] 1/3 present. Missing: ['reject_description', 'reject_type']
2025-04-05 12:00:00 [WARNING] Missing columns (12): ['hcp_id', 'record_type', 'other_coverage', ...]
2025-04-05 12:00:00 [INFO] Feature engineering on 85,432 claims...
2025-04-05 12:00:00 [WARNING] Skipping 'pharmacy_mail_order_pct' — missing mail_order
2025-04-05 12:00:00 [WARNING] Skipping 'prescriber_specialty_valid' — missing prescriber_specialty

Vendor Format Profiles

Profile	Description
auto	Scan file header, detect best match (default)
gsk_iqvia	Expect IQVIA-style column names
generic_csv	Expect generic lower-case names
cms_dmr	CMS Drug Monitoring Report format
elaad_apld	ELAAD/APLD format with MEMBER_ID, HCP_ID, etc.
unknown	No preconceptions, rely fully on synonym matching

Adding a New Vendor Synonym

No code changes needed. Edit config.py → COLUMN_SYNONYMS:

"patient_id": [
    "IQVIA_PATIENT_ID", "PATIENT_ID", "PAT_ID", "MEMBER_ID",
    "NEW_VENDOR_PATIENT_ID",  # ← add your vendor's name here
],

---

Input Data Schema

The pipeline accepts any of the following formats and auto-discovers columns:

Format	Extension	Auto-detect?
CSV	.csv	Yes
Tab-separated	.txt, .tsv	Yes (scans first line for \t)
Gzipped tab	.txt.gz, .csv.gz	Yes
ZIP archive	.zip	Yes (extracts first CSV/TXT inside)
Excel	.xlsx, .xls	Yes (requires openpyxl/xlrd)

Key column groups the pipeline looks for:

• Patient: IQVIA_PATIENT_ID / PATIENT_ID / MEMBER_ID → patient_id
• HCP: IQVIA_PRESCRIBER_ID / HCP_ID / PRESCRIBER_ID / DOCTOR_ID → prescriber_npi
• Claim: CLAIM_NUMBER / CLAIM_NUM / CLAIMID → claim_number
• Drug: NDC / DRUG_NDC / NATIONAL_DRUG_CODE → drug_ndc
• Financial: COPAY_AFTER_BENEFIT / COPAY_AFTER / OOP_COST → copay_after
• Pharmacy: PHARMACY_NABP_NUMBER / PHARMACY_ID / STORE_ID → pharmacy_nabp
• Insurance: PRIMARY_PAYER_BIN / PAYER_BIN / BIN → primary_payer_bin
• Reject: REJECT_CODE / REJECTION_CODE / DENIAL_CD → reject_code

See config.py::COLUMN_SYNONYMS for the full list of 100+ synonyms.

---

35 Business Rules

Original 23 Rules (v3)

#	Rule	Condition	Fraud Signal
1	Early Refill	days_between_fills < 23	Early Refill Abuse
2	Impossible Qty	quantity != 1	Data Error / Fraud
3	Wrong Days Supply	days_supply != 30	Data Error / Fraud
4	Govt Insurance	insurance_type == Government	Program Violation
5	Underage	patient_age < 18	Program Violation
6	Duplicate	Same patient + date + pharmacy	Duplicate Billing
7	NDC Switch	patient_ndc_count > 1	Strength Switching
8	Suspicious Specialty	Prescriber not in valid list	Prescriber Collusion
9	Multi-Program	unique_programs_per_patient > 1	Card Stacking
10	Excessive Fills (90d)	patient_fill_count_90d > 4	Stockpiling
11	High-Risk Reject	reject_code in {76, 88, 79}	Maximizer / DUR
12	Maximizer Cap	maximizer_reject == 1	Benefit Exhaustion
13	Paper Submission	paper_submission == 1	Submission Fraud
14	Plan Switch	plan_switch_flag == 1	Plan Switching
15	Linked Claim	has_linked_claim == 1	Reversal / Adjustment
16	HCP High Benefit	hcp_avg_benefit_per_patient > 500	Prescriber-driven extraction
17	HCP One-Done Concentration	hcp_one_and_done_pct > 0.6	HCP with hit-and-run patients
18	Pharmacy Fraud Risk	pharmacy_fraud_risk_score > 0.6	Composite pharmacy ring score
19	Pharmacy HCP Concentration	pharmacy_hcp_concentration > 0.5	Single HCP dominates pharmacy
20	Pharmacy One-Done	pharmacy_one_and_done_pct > 0.5	Pharmacy with churn-and-burn
21	Short Active Burst	patient_active_duration <= 14 AND total_fills > 1	Quick-fire multi-fill scheme
22	Cross-State	patient_state != pharmacy_state	Out-of-state fraud
23	New Patient Burst	days_since_first <= 7 AND total_fills > 1	Same-week multiple fills

v4 Group-Aware Rules (NEW)

#	Rule	Condition	Fraud Signal
24	Scenario Not Covered	scenario_not_covered_flag == 1	Cash/Rejected under Group 8200
25	Benefit Cap Exceeded	excess_payment_amount > 0	Benefit > allowed cap for group
26	Invalid Period Benefit	invalid_period_benefit_flag == 1	$500 cap used outside Jan-Mar 2024
27	Annual Fill Limit	annual_fill_count > 12	Exceeds 12 fills/year
28	Annual DS Limit	annual_days_supply_count > 360	Exceeds 360 days supply/year
29	Non-Covered NDC	non_covered_ndc_flag == 1	NDC not in covered list
30	Govt With Benefit	govt_claim_with_benefit_flag == 1	Govt plan receiving benefit
31	Quantity Out of Range	quantity < 1 or quantity > 3	Impossible quantity
32	Days Supply Out of Range	days_supply < 1 or days_supply > 90	Invalid supply duration
33	Max Benefit Repeat	Patient+pharmacy hits cap ≥3 times	Organized cap-maximization ring
34	High Cap Utilization	cap_utilization_ratio > 1.0	Paid more than 100% of cap
35	Group Benefit Mismatch	group_benefit_mismatch_flag == 1	Any group+scenario+period mismatch

Rules whose source columns are missing are silently skipped (count = 0).

---

Model: Isolation Forest

IsolationForest(
    n_estimators=200,
    contamination=0.03,        # Adjustable via CLI (use 0.40 for synthetic test data)
    max_samples="auto",
    max_features=1.0,
    bootstrap=False,
    random_state=42,
    n_jobs=-1,
)

Training strategy: Train ONLY on claims where rule_flag == 0 (rule-clean). Score ALL claims.

Degraded mode: If zero rule-clean claims exist (e.g., synthetic data with 40% fraud), the model trains on ALL claims with contamination=min(orig×3, 0.5). This is a safety fallback — real production data will always have rule-clean claims.

Priority score:

priority_score = 0.50 * if_anomaly_score + 0.30 * rule_severity + 0.20 * rule_flag

Risk Tiers:

Tier	Score	Action
Low	0.0–0.3	No action
Medium	0.3–0.6	Monitor
High	0.6–0.8	Investigate (investigation queue)
Critical	0.8–1.0	Immediate investigation + audit trail

---

Outputs

EDA Outputs

File	Type	Description
eda/00_data_quality.json	JSON	Missing values, cardinality, numeric stats, outlier counts
eda/01_temporal_analysis.png	PNG	Monthly volume, day-of-week, quarterly, benefit trends
eda/01_temporal_analysis.json	JSON	Monthly counts + benefit totals
eda/02_financial_analysis.png	PNG	Distribution + boxplot for benefit, copay, OOP, usual customary
eda/02b_benefit_by_scenario.png	PNG	Benefit amount boxplot by claim scenario
eda/02c_benefit_by_group.png	PNG	Benefit amount boxplot by group
eda/03_categorical_analysis.png	PNG	Top 15 bar charts: specialty, scenario, NDC, pharmacy type, group, insurance
eda/03_categorical_analysis.json	JSON	Full top-20 category counts
eda/04_geographic_analysis.png	PNG	Top 20 state distributions + cross-state heatmap
eda/04_geographic_analysis.json	JSON	State counts + cross-state claim count
eda/05_correlation_analysis.png	PNG	Lower-triangle correlation heatmap
eda/05_correlation_analysis.json	JSON	Top 30 correlated feature pairs
eda/06_hierarchical_analysis.png	PNG	Claims/patient, patients/HCP, HCPs/pharmacy, pharmacies/patient
eda/06_hierarchical_analysis.json	JSON	Mean/median/max/std for each hierarchy
eda/07_feature_engineered_eda.png	PNG	Cap utilization, early refill, excess payment, scenario not covered

Core Outputs

File	Type	Description
investigation_queue_top500.csv	CSV	Top 500 highest-priority claims for manual review
scored_claims_full.csv	CSV	All claims with scores + risk tiers
feature_importance_shap.csv	CSV	SHAP ranking of features
metrics.json	JSON	All evaluation metrics + hierarchical summary counts
schema_report.json	JSON	Column mapping audit trail

Hierarchical Summaries

File	Type	Description
transaction_level_summary.csv	CSV	Per-claim analytical view
hcp_level_summary.csv	CSV	Per-HCP investigative lens
pharmacy_level_summary.csv	CSV	Per-pharmacy investigative lens
patient_level_summary.csv	CSV	Per-patient behavioral view

Visualizations

File	Type	Description
01_evaluation_metrics.png	PNG	Score distribution, ROC, PR, tier counts
02_shap_summary.png	PNG	SHAP beeswarm (top 20 features)
03_shap_bar_importance.png	PNG	SHAP bar chart
04_rule_breakdown.png	PNG	Flagged claims by each rule
05_risk_tier_distribution.png	PNG	Tier counts + fraud rates
06_hcp_summary.png	PNG	HCP risk score distribution + benefit vs risk scatter
07_pharmacy_summary.png	PNG	Pharmacy risk score distribution + fraud risk vs priority
08_patient_summary.png	PNG	Patient risk score distribution + one-and-done rate

Model Artifacts

File	Type	Description
model/isolation_forest_model.pkl	PKL	Trained model
model/scaler.pkl	PKL	StandardScaler
model/encoder.pkl	PKL	OrdinalEncoder
model/feature_names.pkl	PKL	Feature name list

---

Configuration: Drug-Agnostic

Edit config.py to configure for any GSK product:

PRODUCT_CONFIG = {
    "product_name": "Trelegy Ellipta",
    "days_supply_expected": 30,
    "quantity_expected": 1,
    "ndc_list": {
        "00173089314": {"strength": "100/62.5/25", "indication": "COPD/Asthma"},
        "00173088714": {"strength": "200/62.5/25", "indication": "Asthma"},
    },
    "valid_prescriber_specialties": ["Pulmonology", "Allergy/Immunology", ...],
    "suspicious_prescriber_specialties": ["Dermatology", "Orthopedics", ...],
    "early_refill_threshold_days": 23,
    "max_fills_90d": 4,
    # HCP thresholds
    "hcp_high_benefit_threshold": 500.0,
    "hcp_patient_concentration_threshold": 0.6,
    # Pharmacy thresholds
    "pharmacy_high_benefit_threshold": 450.0,
    "pharmacy_hcp_concentration_threshold": 0.5,
    "pharmacy_one_done_threshold": 0.5,
    # Patient thresholds
    "patient_gap_short_threshold": 15,
    "patient_gap_long_threshold": 60,
    "patient_max_active_duration_days": 180,
    # Group-aware limits
    "annual_max_fills_per_patient": 12,
    "annual_max_days_supply_per_patient": 360,
    "quantity_min": 1,
    "quantity_max": 3,
    "days_supply_min": 1,
    "days_supply_max": 90,
    ...
}

---

Tech Stack

• Python 3.9+
• pandas ≥ 2.0.0, numpy ≥ 1.24.0, scikit-learn ≥ 1.3.0
• shap ≥ 0.42.0, matplotlib ≥ 3.7.0, seaborn ≥ 0.12.0, joblib ≥ 1.3.0, pyarrow ≥ 12.0.0

---

License

Proprietary — GSK internal use.

Generated by ML Intern

This model repository was generated by ML Intern, an agent for machine learning research and development on the Hugging Face Hub.

• Try ML Intern: https://smolagents-ml-intern.hf.space
• Source code: https://github.com/huggingface/ml-intern

Usage

from transformers import AutoModelForCausalLM, AutoTokenizer

model_id = 'Harsh2396/gsk-copay-fraud-detection'
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(model_id)

For non-causal architectures, replace AutoModelForCausalLM with the appropriate AutoModel class.