Voucher Abuse Detection with Prompt-based Fine-tuning on Graph Neural Networks

TL;DR

This paper introduces VPGNN, a prompt-based fine-tuning framework for GNNs that improves voucher abuse detection by reducing the objective gap between pre-training and downstream tasks, showing significant performance gains.

Contribution

The paper proposes a novel graph prompting function that reformulates voucher abuse detection as a similar task to pre-training, enhancing GNN fine-tuning effectiveness.

Findings

VPGNN outperforms existing models in few-shot and semi-supervised settings.

Online deployment shows a 23.4% accuracy improvement.

Framework effectively narrows the objective gap between pre-training and downstream tasks.

Abstract

Voucher abuse detection is an important anomaly detection problem in E-commerce. While many GNN-based solutions have emerged, the supervised paradigm depends on a large quantity of labeled data. A popular alternative is to adopt self-supervised pre-training using label-free data, and further fine-tune on a downstream task with limited labels. Nevertheless, the "pre-train, fine-tune" paradigm is often plagued by the objective gap between pre-training and downstream tasks. Hence, we propose VPGNN, a prompt-based fine-tuning framework on GNNs for voucher abuse detection. We design a novel graph prompting function to reformulate the downstream task into a similar template as the pretext task in pre-training, thereby narrowing the objective gap. Extensive experiments on both proprietary and public datasets demonstrate the strength of VPGNN in both few-shot and semi-supervised scenarios. Moreover, an online deployment of VPGNN in a production environment shows a 23.4% improvement over two existing deployed models.