Improving Micro-video Recommendation via Contrastive Multiple Interests

TL;DR

This paper introduces CMI, a contrastive learning-based model that extracts multiple user interests from micro-video interaction data, improving personalized recommendations without relying on expensive multi-modal information.

Contribution

The paper proposes a novel contrastive multi-interest learning approach that decouples user interests using orthogonal categories, enhancing recommendation accuracy.

Findings

CMI outperforms existing baselines on two datasets.

It effectively captures multiple user interests.

The model improves robustness of interest embeddings.

Abstract

With the rapid increase of micro-video creators and viewers, how to make personalized recommendations from a large number of candidates to viewers begins to attract more and more attention. However, existing micro-video recommendation models rely on expensive multi-modal information and learn an overall interest embedding that cannot reflect the user's multiple interests in micro-videos. Recently, contrastive learning provides a new opportunity for refining the existing recommendation techniques. Therefore, in this paper, we propose to extract contrastive multi-interests and devise a micro-video recommendation model CMI. Specifically, CMI learns multiple interest embeddings for each user from his/her historical interaction sequence, in which the implicit orthogonal micro-video categories are used to decouple multiple user interests. Moreover, it establishes the contrastive multi-interest loss to improve the robustness of interest embeddings and the performance of recommendations. The results of experiments on two micro-video datasets demonstrate that CMI achieves state-of-the-art performance over existing baselines.