Addressing Noise and Stochasticity in Fraud Detection for Service Networks

Wenxin Zhang 1 , Ding Xu 2 , Xi Xuan 3 , Lei Jiang 1 , Guangzhen Yao 4 , Renda Han 5 , Xiangxiang Lang 5 , Cuicui Luo 1 · 2025

Read Addressing Noise and Stochasticity in Fraud Detection for Service Networks by Wenxin Zhang 1 , Ding Xu 2 , Xi Xuan 3 , Lei Jiang 1 , Guangzhen Yao 4 , Renda Han 5 , Xiangxiang Lang 5 , Cuicui Luo 1 (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2505-00946.txt.

Description

Wenxin Zhang and co-authors clean up fraud detection on service networks. Their 2025 paper notes that graph-based detectors read signals at different frequencies to spot fraud. Noise picked up as signals spread muddies them, though, and blending the frequencies erases what makes each one useful.

Their network answers both flaws. It splits the graph into like-minded and mixed neighborhoods to separate the frequencies cleanly. Information bottleneck theory then squeezes each encoding down to its telling features. Prototype learning fuses the streams while keeping their distinct characters. Across three real data sets, the system beat the leading detectors. The authors present the design as a general fix for noisy graph signals, with fraud as the proving ground.

Catalog record

CREATORWenxin Zhang 1 , Ding Xu 2 , Xi Xuan 3 , Lei Jiang 1 , Guangzhen Yao 4 , Renda Han 5 , Xiangxiang Lang 5 , Cuicui Luo 1
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE50 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2505-00946

Browse the full library · About the Tower of Babel