DMBGN: Deep Multi-Behavior Graph Networks for Voucher Redemption Rate Prediction

TL;DR

This paper introduces DMBGN, a graph neural network model that captures complex user-voucher-item relationships and user behavior patterns to improve voucher redemption rate prediction in e-commerce, outperforming existing models.

Contribution

The paper proposes a novel Deep Multi-behavior Graph Network that models intricate user-voucher-item relations and user behaviors before and after voucher collection for better prediction accuracy.

Findings

DMBGN achieves 10-16% relative AUC improvement over DNN.

DMBGN outperforms DIN with 2-4% AUC improvement.

Extensive experiments validate the effectiveness of the proposed model.

Abstract

In E-commerce, vouchers are important marketing tools to enhance users' engagement and boost sales and revenue. The likelihood that a user redeems a voucher is a key factor in voucher distribution decision. User-item Click-Through-Rate (CTR) models are often applied to predict the user-voucher redemption rate. However, the voucher scenario involves more complicated relations among users, items and vouchers. The users' historical behavior in a voucher collection activity reflects users' voucher usage patterns, which is nevertheless overlooked by the CTR-based solutions. In this paper, we propose a Deep Multi-behavior Graph Networks (DMBGN) to shed light on this field for the voucher redemption rate prediction. The complex structural user-voucher-item relationships are captured by a User-Behavior Voucher Graph (UVG). User behavior happening both before and after voucher collection is taken into consideration, and a high-level representation is extracted by Higher-order Graph Neural Networks. On top of a sequence of UVGs, an attention network is built which can help to learn users' long-term voucher redemption preference. Extensive experiments on three large-scale production datasets demonstrate the proposed DMBGN model is effective, with 10% to 16% relative AUC improvement over Deep Neural Networks (DNN), and 2% to 4% AUC improvement over Deep Interest Network (DIN). Source code and a sample dataset are made publicly available to facilitate future research.