The objective of this project is to develop a system capable of detecting fraudulent or suspicious-looking credit card transactions in real-time in a production environment. This project aims to provide a model that can predict on a per transaction basis whether the transaction seems suspicious or not, depending upon both the transaction independent of the user and corroborated by the user’s individual purchasing behaviour or history.
This project aims to provide a way to identify fraudulent transactions from a forementioned data. Mitigation of fraud is left to the discretion of the providers/lenders themselves. A variety of options are available for this, such as manual verification and so forth.