I created this application for Western Governors University's Computer Science capstone project.
This application was created for a fictitious credit card company called IOU Inc. IOU used analysts instead of computers to detect fraud before this application. This application automated that process. It uses the random forest classifier, a machine learning algorithm, to categorize transactions as either fraudulent or legitimate. The program's target user is the management staff in IOU's fraud detection department. Once a transaction has been classified as fraudulent, it will appear on the user's dashboard tab. At that point, the user may assign it to an employee for further review. In the fictional scenario of the project IOU already has an application for employees to review transactions in detail. Therefore, this was beyond the scope of the project.
For more information about the features of the application, please refer to User_Guide.docx in the Documentation folder. Application_Files.docx, which is also in the Documentation folder, contains a description of each file.
How to install and use the application: Please refer to User_Guide.docx in the Documentation folder for instructions.
The dataset used was collected and analyzed during a research collaboration of Worldline and the Machine Learning Group (http://mlg.ulb.ac.be) of ULB (Université Libre de Bruxelles) on big data mining and fraud detection. I obtained it from https://www.kaggle.com/mlg-ulb/creditcardfraud.