Kar Chan Undergraduate Dissertation 2017/18

Credit Card Fraud detection using Machine Learning in TensorFlow

Supervised by M.Alvarez_Lopez

Abstract

Credit card payment has always been the most widely used payment method since decades. With the advancement of digital technologies, credit card payment has simply evolved from paying by inserting your chip card to paying online using the virtual credit card in your phone. Unfortunately, technology like this is not 100 percent theft proof. Hence, card companies must always keep track on your transactions to make sure that those are legitimate transactions. With the help of Machine Learning, card companies are now able to detect fraud transactions by looking at the patterns and anomalies of the credit card data.

The aim of the project is to determine the best Machine Learning technique that is suitable for credit card fraud detection. Different kinds of learning algorithm will be implemented together with methods that handle the imbalanced credit card data and the best solution is selected.