/Credit_Card-Fraud_Detection-AutoEncoders-XGB

Credit card Fraud Detection using AutoEncoders Neural network to encode complete dataset. Training the Genuine Transaction Data alone and create Anomaly Detecting Classification Model using the simple Logistic Regression Classifier and XGBoost to predict the Fraud transaction from the Encoded dataset.

Primary LanguageJupyter Notebook

Credit_Card-Fraud_Detection-AutoEncoders-XGB

Credit card Fraud Detection using the Dataset from Kaggle

Useful Links

This is link to the dataset in Kaggle - https://www.kaggle.com/mlg-ulb/creditcardfraud

Autoencoders - https://towardsdatascience.com/auto-encoder-what-is-it-and-what-is-it-used-for-part-1-3e5c6f017726

XGBoost - https://www.kaggle.com/stuarthallows/using-xgboost-with-scikit-learn

Ultimately, we can use many other classification and Boosting techniques on the Encoded dataset (using the Autoencoder) and we can make use of the Ensembling techniques to improvise the accuracy.