Fraud Detection-using-Resampling

Dependencies

• python3.6
• pandas
• sklearn
• matplotlib
• imblearn

Running Steps

Update:

1. TRAINFILENAME: The CSV file of training data
2. TEST_FR: Fraction of training data to be kept aside for test data
3. DISC_FEATURES_COL_TO_USE: Python List of Column names containing Discrete Features in TRAINFILENAME
4. CONT_FEATURES_COL_TO_USE: List of Column names containing Continuous Features in TRAINFILENAME
5. DISC_TARGET_COL_TO_USE: Name of Column names containing Target variable in TRAINFILENAME
6. Type arg passed to training function in main(). Type of model to be used from:
   • LR: Logistic Regression
   • SVM: Support Vector Classifier
   • RF: Random forest
   • GBC: Gradient Boosting Classifier
   • Default: Naive Bayes

Run: As ussual python script