Our research is inspired and rooted around the application of machine learning techniques to model users’ behavior from historical data in order to detect frauds as suspicious deviations from the models built.
Main research outcomes:
BankSealer: a decision support system for online banking fraud analysis and investigation. During a training phase, BankSealer builds easy-to-understand models for each customer’s spending habits, based on historical transactions. At runtime, BankSealer supports analysts by ranking new transactions that deviate from the learned profiles, with an output that has an easily understandable, immediate statistical meaning.
Application of supervised learning technique based on Multi-Objective Genetic Algorithm that exploit analyst’s feedbacks for tuning BankSealer’s parameters to automatize the feature weighting task and to improve detection performances (i.e., push real frauds higher in the ranking).