CreditCardFraudDetection

The purpose of this project is to find fraudulent transactions in the credit card dataset, based on 30 features provided. The features contain random numeric values ranging between -1 to 1, so it is important to determine which features are important and how to correctly classify them.

Tasks performed in this notebook:

• Reading Data
• Feature Importance Calculation
  • Plot of FScores vs Features
    • Extreme Gradient Boosting (XGB) Classifier
    • Extra Trees Classifier
  • Selecting Top Features
• Training and Classification
  • SVM
  • Random Forest Classifier
• Imbalanced Classes

Dataset available at https://www.kaggle.com/mlg-ulb/creditcardfraud