BTG: A Bridge to Graph machine learning in telecommunications fraud detection

Xinxin Hu, Hongchang Chen, Shuxin Liu, Haocong Jiang, Guanghan Chu, Ran Li

Abstract: Highlights•We propose a new end-to-end telecom fraud detection framework.•Our method bridges the gap between sparse connectivity data and graph machine learning.•An effective method for extracting user multi-model behavior characteristics.•A publicly available real-world Call Detail Records (CDR) dataset shared with peers.