The study introduces an inductive link‑prediction framework for detecting payroll fraud on synthetic HR/ERP graphs, comparing analyst rules, path‑feature models, and graph neural networks (GNNs). A gradient‑boosted model using chained‑join features achieved the highest link average precision (AP 0.823), outperforming a GraIL‑style GNN (AP 0.663) and demonstrating that multi‑hop gains are not exclusive to GNNs. In additional experiments, the chained‑join model again led on a new cohort (AP 0.798), while the relational GNN surpassed record‑aware rules (AP 0.707 vs. 0.504) under various conditions, highlighting the distinct roles of multi‑hop discovery, record‑based discrimination, and review prioritization.
arXiv:2607. 19266v1 Announce Type: cross Abstract: Fraud detection systems must scale with rising transaction volume while remaining explainable and reviewable.
SemHGT is a new framework that enhances heterogeneous graph learning for financial fraud detection by explicitly incorporating business semantics of structured attributes. It uses role‑aware prompts to generate risk‑related descriptions, maps these to semantic vectors, and then guides multi‑round sampling of transaction paths based on semantic similarity. The framework combines local path evidence and global neighbor summaries in a unified transformer, achieving superior performance on three benchmark datasets compared to existing baselines.
arXiv:2609.14234v1 Announce Type: cross Abstract: Financial fraud in corporate transaction networks has grown more coordinated and harder to detect with rule-based engines and with classical learning...
arXiv:2607. 11212v1 Announce Type: new Abstract: Relational fraud detection can exploit both label-free graph context and label-derived neighborhood evidence, but these two information sources obey different validity conditions.
arXiv:2610.27107v1 Announce Type: new Abstract: Credit card fraud is a significant problem for financial institutions, resulting in substantial monetary loss each year. Fraud patterns and methods are...
Evi-VN is a new framework for graph neural network (GNN) fraud detection that injects evidence from diverse modalities—structured records, text, images, and audio—into virtual node representations. It targets shared hard regions where standard GNNs fail to distinguish well‑disguised fraudsters from legitimate users, correcting blind spots without disrupting reliable predictions. Experiments on bot, fake‑review, refund‑evidence, and telecom‑fraud datasets demonstrate its effectiveness across generic, fraud‑specific, and unseen GNNs.
The paper presents a graph‑based method for detecting money‑mule accounts by treating a bank’s relational schema as a heterogeneous graph and scoring each account using multi‑hop neighbourhoods, per‑column encoders, relative‑time encoding, and HeteroGraphSAGE message passing. On a five‑year RBIH retail‑banking dataset of 400 million transactions, the model achieves 83 % precision and a 0.5 % false‑alarm rate, recovering nine out of ten held‑out mules and outperforming tree‑based baselines in recall while maintaining precision deep into the alert queue. The approach can be deployed directly on a bank’s existing database without constructing a custom feature graph.
Credit card fraud is a significant problem for financial institutions, resulting in substantial monetary loss each year. Fraud patterns and methods are continuously evolving, with fraudulent transacti...
Money-mule networks move criminal funds through many small transfers, each of which looks ordinary on its own. What marks an account as a mule lies outside any single transfer: in the counterparties a...
arXiv:2601. 11073v3 Announce Type: replace-cross Abstract: Online financial services constitute an essential component of contemporary web ecosystems, yet their openness introduces substantial exposure to fraud that harms vulnerable users and weakens trust in digital finance.
arXiv:2608. 15177v1 Announce Type: cross Abstract: The increasing complexity of digital financial systems has reshaped financial fraud detection from isolated transaction classification into relational risk reasoning over interconnected financial entities.
arXiv:2609.07100v1 Announce Type: cross Abstract: Credit card fraud detection typically relies on tabular features, while repeated attributes can also provide useful relational signals. This paper pr...