Payments Fraud and Anomaly Detection

---

Using unsupervised and supervised learning methods to detect fraud and anomalies in credit card data. When all you know are simple transactiondetails of a purchase

---

Datasets:

1. Credit Card Fraud (labeled)
   • This dataset contains credit card transactions made by European cardholders over a period of two days in September 2013. It has a total of 284,807 transactions, out of which 492 are fraud.
2. Chicago Vendor Payments
   • Public data showing all vendor payments made by the City of Chicago from 1996 to present

---

Project Structure

├── data
│   ├── holds raw and cleaned data
├── notebooks
│   ├── EDA and preprocessing
│   ├── Model building and training
├── src
│   ├── preprocessing
│   ├── modeling

• The top folders hold a structure similar to above, but each with a focus
  • fraud-detection focuses on the labeled data and uses machine learning models to determine if a purchase is fraudulent
  • anomaly-detecton uses unsupervised learning on the Chicago dataset to find anomalies in the vendor payments
• The data directory contains the raw and cleaned datasets
• The notebooks directory contains Jupyter notebooks for performing exploratory data analysis (EDA), preprocessing, and modeling.
• The src directory contains Python modules for preprocessing and modeling.