Fraudulent E-Commerce Transactions (2)

Data Loading and Data Preparation

Data Understanding and Visualization

Data Preprocessing

Training the Autoencoder

Optimizing threshold K

Final Performance

Keras Autoencoder Architecture

The Fraud Detector
- data normalization
- reproduction of data by autoencoder
- MSE calculation
- Applying threshold

Taking Actions

IF trx is legitimate => do nothing
IF trx is fraud candidate => send email to owner

Deployment

Data Modeling

Modeling Batch

Select the Product Category to train the model