MV-HAN: A Hybrid Attentive Networks based Multi-View Learning Model for Large-scale Contents Recommendation

TL;DR

This paper introduces MV-HAN, a hybrid attentive network model for multi-view content recommendation that enhances retrieval accuracy and efficiency by leveraging high-order feature interactions and knowledge transfer across data types.

Contribution

The paper presents a novel multi-view hybrid attentive network model that improves large-scale content retrieval by enabling high-order feature interactions and effective knowledge sharing.

Findings

Significantly outperforms baseline models in offline content retrieval tasks.

Achieves notable improvements in online recommendation quality during real-world deployment.

Enhances retrieval speed through a two-tower model and approximate nearest neighbor search.

Abstract

Industrial recommender systems usually employ multi-source data to improve the recommendation quality, while effectively sharing information between different data sources remain a challenge. In this paper, we introduce a novel Multi-View Approach with Hybrid Attentive Networks (MV-HAN) for contents retrieval at the matching stage of recommender systems. The proposed model enables high-order feature interaction from various input features while effectively transferring knowledge between different types. By employing a well-placed parameters sharing strategy, the MV-HAN substantially improves the retrieval performance in sparse types. The designed MV-HAN inherits the efficiency advantages in the online service from the two-tower model, by mapping users and contents of different types into the same features space. This enables fast retrieval of similar contents with an approximate nearest neighbor algorithm. We conduct offline experiments on several industrial datasets, demonstrating that the proposed MV-HAN significantly outperforms baselines on the content retrieval tasks. Importantly, the MV-HAN is deployed in a real-world matching system. Online A/B test results show that the proposed method can significantly improve the quality of recommendations.