Financial Applications of Graph Neural Networks: A Review


Creative Commons License

Alptekin G.

International Conference of TechInternational Conference of Technology Analysis, Fintech and Financial Services (TAFFS 2022)nology Analysis, Fintech and Financial Services (TAFFS 2022), Kolkata, Hindistan, 11 - 13 Mart 2022, ss.1-4

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Basıldığı Şehir: Kolkata
  • Basıldığı Ülke: Hindistan
  • Sayfa Sayıları: ss.1-4
  • Galatasaray Üniversitesi Adresli: Evet

Özet

Graph Neural Networks (GNNs) is one of the deep learning methods, that is designed to study on data structured with graphs. In this paper, we will present several recent works on the applications of graph neural networks (GNN) on financial domain. Card fraud, forecasting in bitcoin operations, fraud detection, linking bank customers, finding credit rating scores, electronic commerce and transactions management are the main research areas in the field of finance. State-of-the-art methods apply machine learning-based approaches to detect fraudulent behavior from transaction records. However, they may not be appropriate for real life’s large fraud detection systems with complex and unpredictable relations. GNN aims at coping with the larger sizes and more complex relationships.