Fraud-detection-PaySim

Classifying normal and fraudulent transactions on synthetic dataset generated with PaySim simulator. Various models are trained to classify fraudulent/regular transactions by tuning the class weights to deal with imbalanced dataset.

Models used:

• Logistic Regression
• Linear SVM
• SVM with RBF Kernel

Data Source:

• Kaggle : https://www.kaggle.com/ntnu-testimon/paysim1