TRAVELFRAUDBENCH: A Configurable Evaluation Framework for GNN Fraud Ring Detection in Travel Networks
Bhavana Sajja

TL;DR
TravelFraudBench (TFG) is a configurable benchmark for evaluating GNNs on travel fraud ring detection, simulating diverse fraud topologies in a heterogeneous graph with comprehensive evaluation metrics.
Contribution
Introduces TFG, a novel, configurable benchmark for travel fraud detection that evaluates GNNs across multiple fraud ring topologies and provides open-source datasets and tools.
Findings
GraphSAGE achieves near-perfect AUC of 0.992 in fraud detection.
Graph structure significantly improves detection performance over baseline.
Device and IP co-occurrence are primary signals for fraud rings.
Abstract
We introduce TravelFraudBench (TFG), a configurable benchmark for evaluating graph neural networks (GNNs) on fraud ring detection in travel platform graphs. Existing benchmarks--YelpChi, Amazon-Fraud, Elliptic, PaySim--cover single node types or domain-generic patterns with no mechanism to evaluate across structurally distinct fraud ring topologies. TFG simulates three travel-specific ring types--ticketing fraud (star topology with shared device/IP clusters), ghost hotel schemes (reviewer x hotel bipartite cliques), and account takeover rings (loyalty transfer chains)--in a heterogeneous graph with 9 node types and 12 edge types. Ring size, count, fraud rate, scale (500 to 200,000 nodes), and composition are fully configurable. We evaluate six methods--MLP, GraphSAGE, RGCN-proj, HAN, RGCN, and PC-GNN--under a ring-based split where each ring appears entirely in one partition,…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Code & Models
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
