Temporal graph transformer improves credit card fraud risk ranking
Temporal Heterogeneous Graph Transformer for Credit Card Fraud Detection
Machine LearningArtificial Intelligence
Summary
Credit card fraud detection usually looks at simple facts about each transaction, but repeated details across transactions can also help spot fraud. The authors created a new method using a type of AI called a Transformer that looks at transactions and their relationships over time. This method uses special timing signals to understand when transactions happened. Their tests showed this method can help rank fraud risks well, though it did not outperform some existing techniques. This work points to the usefulness of considering relationships between transactions for better fraud detection.
What this means in practice
- •For fraud analysts: Use relational and temporal graph features alongside transaction data to better prioritize investigation of fraudulent credit card transactions.
- •For financial software engineers: Incorporate temporal heterogeneous graph transformers to enhance risk scoring models for credit card transactions by capturing relation patterns.
Tested on one dataset.
Authors
Qinwen Yan
Abstract
Credit card fraud detection typically relies on tabular features, while repeated attributes can also provide useful relational signals. This paper proposes THGT-FD, a Temporal Heterogeneous Graph Transformer for Fraud Detection. Each transaction is represented using one transaction token and six types of relation tokens and incorporates Time2Vec encoding into the transaction representation. A Transformer learns the interactions among these tokens within each individual transaction and then outputs a fraud probability. Experiments were conducted on 150,000 transactions sampled from the IEEE-CIS Fraud Detection dataset and chronologically partitioned according to TransactionDT. On the test set, THGT-FD achieved an AUC-ROC of 0.8536, an average precision of 0.4164, and a Recall@5% of 0.4708. The class-weighted histogram-based gradient-boosting baseline achieved an AUC-ROC of 0.8722. The results indicate that relation tokens provide useful information for fraud-risk ranking, although the current model does not yet incorporate entity-level historical aggregation.