README.md

Credit-Card-Fraud-Detection

This Study/Project deals with predicting fraudulent credit card transactions. The data is taken from famous Kaggle dataset: https://www.kaggle.com/mlg-ulb/creditcardfraud Since the data is highly imbalanced, we have approcahed it in following ways;

1. Random Forest Classification with class weight.
2. Random Forest Classification with under sampling.
3. Random Forest Classification with over sampling.

We have also used unsupervised ML algorithms for anomaly detection.

1. Local Outlier Factor
2. Isolation Forest

We have tried to understand different model evaluation measures such as accuracy, confusion matrix, precision, recall and f1 score.