Credit-Card-Fraud-Prediction

Credit card fraud is a wide-ranging term for theft and fraud committed using or involving a payment card, such as a credit card or debit card, as a fraudulent source of funds in a transaction. I have used credit card fraud dataset obatined from kaggle (https://www.kaggle.com/mlg-ulb/creditcardfraud ) to detect credit card fraud.

1. Exploratory data analysis.
2. Split data into train and test. Normalise Data.
3. Train a MLP: a. With two hidden layers (8nodes and 4nodes) and one output layer (2 nodes) b. Relu activation c. Adam Optimizer
4. Train Autoencoder-Decoder a. Encoder of 8 nodes b. tanh activation c. RMSProp Optimiser d.Use auc score to validate the result