Fraudulent Transaction Detection is an innovative project designed to identify potential fraudulent transactions. By analyzing various factors such as average transaction amount per day, total declined transactions, high-risk country transactions, and foreign country transactions, this system utilizes the powerful Isolation Forest algorithm to accurately detect fraudulent activities.
This project offers two types of inputs for user convenience. The first option involves providing a .csp file, where the system processes the data and generates an output .csp file indicating the presence of fraud. Alternatively, users can manually enter the required parameters to perform fraud detection.