Behavior Sequence Transformer for E-commerce Recommendation in Alibaba

Abstract

Deep learning based methods have been widely used in industrial recommendation systems (RSs). Previous works adopt an Embedding&MLP paradigm: raw features are embedded into low-dimensional vectors, which are then fed on to MLP for final recommendations. However, most of these works just concatenate different features, ignoring the sequential nature of users' behaviors. In this paper, we propose to use the powerful Transformer model to capture the sequential signals underlying users' behavior sequences for recommendation in Alibaba. Experimental results demonstrate the superiority of the proposed model, which is then deployed online at Taobao and obtain significant improvements in online Click-Through-Rate (CTR) comparing to two baselines.

4 pages, 1 figure

References in corpus (6)

• xDeepFM: Combining Explicit and Implicit Feature Interactions for Recommender Systems
• Learning Tree-based Deep Model for Recommender Systems
• Wide & Deep Learning for Recommender Systems
• BERT4Rec: Sequential Recommendation with Bidirectional Encoder Representations from Transformer
• Multi-Interest Network with Dynamic Routing for Recommendation at Tmall
• Personalized Re-ranking for Recommendation

Cited by in corpus (5)

• COLD: Towards the Next Generation of Pre-Ranking System
• Entire Space Multi-Task Modeling via Post-Click Behavior Decomposition for Conversion Rate Prediction
• Sequential Recommendation with Relation-Aware Kernelized Self-Attention
• Scaling Law for Recommendation Models: Towards General-purpose User Representations
• InfoBehavior: Self-supervised Representation Learning for Ultra-long Behavior Sequence via Hierarchical Grouping