9004 Problems of information security. Computer systems Проблемы информационной безопасности. Компьютерные системы 2071-8217 10 10.48612/jisp/reag-mhhb-n41d Banking fraud detection using graph neural networks Обнаружение банковского мошенничества с применением графовых нейронных сетей 0000-0003-0374-4649 Sergadeeva Anastasia nsspbpoly@gmail.com 0000-0003-2849-4682 Lavrova Daria lavrova_ds@spbstu.ru Peter the Great St. Petersburg Polytechnic University 24 12 2021 4 112 122

This paper proposes the application of graph neural networks to detect bank fraud. Financial transactions are represented in the form of a graph, and the use of graph neural networks allows the detection of transactions characteristic of fraudulent schemes. Experimental results demonstrate the promise of the proposed approach

 graph neural networks bank fraud anomaly detection convolutional neural networks information security financial data analysis