Detecting client-side e-banking fraud using a heuristic model

This research proposes and implements a heuristic model to detect client-side e-banking fraud caused by malware. Results show that the model is promising and is able to detect malicious injections from malware. To validate the developed model, an additional experiment is performed in which unknown web pages, adapted by recent malware are correctly classified based on historical, malicious pages of a bank. However, validation of the results with a more representative dataset is required. The classification process of a web page is performed with a mean of 0.176 seconds. Improvement of the developed model may lower impact on resources and execution time. Supervisor: Mark Wiggerman [email protected]