Orthogonal Hyper-category Guided Multi-interest Elicitation for Micro-video Matching

TL;DR

This paper introduces OPAL, a model for micro-video matching that effectively captures multiple user interests through interest disentanglement and orthogonal hyper-category guidance, improving recommendation diversity and accuracy.

Contribution

The paper proposes a novel interest elicitation model with a two-stage training strategy that enhances interest disentanglement and interest evolution modeling in micro-video recommendation.

Findings

OPAL outperforms six state-of-the-art models in recall and hit rate.

OPAL effectively captures diversified user interests.

The model demonstrates strong performance on real-world datasets.

Abstract

Watching micro-videos is becoming a part of public daily life. Usually, user watching behaviors are thought to be rooted in their multiple different interests. In the paper, we propose a model named OPAL for micro-video matching, which elicits a user's multiple heterogeneous interests by disentangling multiple soft and hard interest embeddings from user interactions. Moreover, OPAL employs a two-stage training strategy, in which the pre-train is to generate soft interests from historical interactions under the guidance of orthogonal hyper-categories of micro-videos and the fine-tune is to reinforce the degree of disentanglement among the interests and learn the temporal evolution of each interest of each user. We conduct extensive experiments on two real-world datasets. The results show that OPAL not only returns diversified micro-videos but also outperforms six state-of-the-art models in terms of recall and hit rate.