Hybrid Machine Learning System for Solving Fraud Detection Tasks
Olena А. Vynokurova, Dmytro Peleshko, Oleksandr Fedorovych Bondarenko, Vadim Ilyasov, Vladislav Serzhantov, Marta Peleshko · 2020
In parallel with technological development the problem of fraud detection is becoming more and more important. Increasing number of electronic transactions in various business environments, on the one hand, and software and technology development, on the other hand, lead to an active increase in electronic crime. In the paper the hybrid system of machine learning for solving tasks of anomalies detection has been proposed. This hybrid system consists of two subsystems - anomalies detection subsystem (based on unsupervised learning) and the interpretation subsystem of anomaly type (based on supervised system). The advantage of proposed hybrid system is the high-speed data processing when the data are fed in real time. The effectiveness of the proposed approach was confirmed during the solution of the detecting anomalies problem based on real data streams.