Papers › SplitGNN: Spectral Graph Neural Network for Fraud Detection against Heterophily

SplitGNN: Spectral Graph Neural Network for Fraud Detection against Heterophily

21 Oct 2023CIKM 2023 10archive 2025-07-28

Bin Wu, Xinyu Yao, Boyan Zhang, Kuo-Ming Chao, Yinsheng Li

Fraudsters in the real world frequently add more legitimate links while concealing their direct ones with other fraudsters, leading to heterophily in fraud graphs, which is a problem that most GNN-based techniques are not built to solve. Several works have been proposed to tackle the issue from the spatial domain. However, researches on addressing the heterophily problem in the spectral domain are still limited due to a lack of understanding of spectral energy distribution in graphs with heterophily. In this paper, we analyze the spectral distribution with different heterophily degrees and observe that the heterophily of fraud nodes leads to the spectral energy moving from low-frequency to high-frequency. Further, we verify that splitting graphs using heterophilic and homophilic edges can obtain more significant expressions of signals in different frequency bands. The observation drives us to propose the spectral graph neural network, SplitGNN, to capture signals for fraud detection against heterophily. SplitGNN uses an edge classifier to split the original graph and adopts flexible band-pass graph filters to learn representations. Extensive experiments on real-world datasets demonstrate the effectiveness of our proposed method. The code and data are available at https://github.com/Split-GNN/SplitGNN.

PaperPDFCode

Code

split-gnn/splitgnn officialmentioned in paperpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Fraud DetectionGraph Neural NetworkNode Classification

Datasets

Introduced by this paper, per the archive.

FDCompCN

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Fraud Detection FDCompCN SplitGNN AUC-ROC 68.98 #1 of 1 Archive leaderboard report
Fraud Detection Yelp-Fraud SplitGNN AUC-ROC 92.03 #5 of 10 Archive leaderboard report
Node Classification Yelp-Fraud SplitGNN AUC-ROC 92.03 #4 of 9 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections