Behavior Sequence Transformer for E-commerce Recommendation in Alibaba
This work addresses the need for better sequential modeling in industrial recommendation systems, specifically for e-commerce platforms like Alibaba, and is incremental as it adapts an existing method (Transformer) to a known bottleneck.
The paper tackled the problem of ignoring sequential user behavior in e-commerce recommendations by proposing a Transformer-based model, which achieved significant improvements in online Click-Through-Rate (CTR) compared to baselines when deployed on Taobao.
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.