Counterfactual Learning-Driven Representation Disentanglement for Search-Enhanced Recommendation

TL;DR

This paper introduces a counterfactual learning framework to disentangle item features into query-independent and query-related components, improving search-enhanced recommendation by reducing negative transfer from search-specific features.

Contribution

It proposes a novel counterfactual learning-driven representation disentanglement method that isolates general item features from search-specific features for better recommendation performance.

Findings

Effective in both collaborative filtering and sequential recommendation.

Reduces negative transfer from search-specific features.

Improves recommendation accuracy with disentangled representations.

Abstract

For recommender systems in internet platforms, search activities provide additional insights into user interest through query-click interactions with items, and are thus widely used for enhancing personalized recommendation. However, these interacted items not only have transferable features matching users' interest helpful for the recommendation domain, but also have features related to users' unique intents in the search domain. Such domain gap of item features is neglected by most current search-enhanced recommendation methods. They directly incorporate these search behaviors into recommendation, and thus introduce partial negative transfer. To address this, we propose a Counterfactual learning-driven representation disentanglement framework for search-enhanced recommendation, based on the common belief that a user would click an item under a query not solely because of the item-query match but also due to the item's query-independent general features (e.g., color or style) that interest the user. These general features exclude the reflection of search-specific intents contained in queries, ensuring a pure match to users' underlying interest to complement recommendation. According to counterfactual thinking, how would user preferences and query match change for items if we removed their query-related features in search, we leverage search queries to construct counterfactual signals to disentangle item representations, isolating only query-independent general features. These representations subsequently enable feature augmentation and data augmentation for the recommendation scenario. Comprehensive experiments on real datasets demonstrate ClardRec is effective in both collaborative filtering and sequential recommendation scenarios.