jeffjohannsen/Fraud_Detection

Machine learning fraud detection system with Flask web interface and NLP feature engineering

32
/ 100
Emerging

Combines categorical, numerical, and HTML-formatted event data through a five-step cleaning pipeline (type conversion, null handling, feature removal, nested aggregation, HTML parsing) before training multiple ML models with NLP-derived probabilities from text fields. Implements a dual-threshold risk stratification (≥0.10 for high risk, ≥0.03 for medium risk) optimized to recall 92.9% of fraudulent transactions, with chronological train/test splits preventing data leakage. Flask web interface displays predictions across three risk categories for manual review workflows, targeting the business priority of catching nearly all fraud despite 12.4% false positive rates.

No Package No Dependents
Maintenance 10 / 25
Adoption 4 / 25
Maturity 9 / 25
Community 9 / 25

How are scores calculated?

Stars

7

Forks

1

Language

Jupyter Notebook

License

MIT

Last pushed

Jan 13, 2026

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/nlp/jeffjohannsen/Fraud_Detection"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.