Snippet-based Conversational Recommender System

TL;DR

SnipRec introduces a resource-efficient, snippet-based conversational recommender system that leverages user reviews and large language models to better capture preferences and improve recommendation accuracy across multiple domains.

Contribution

The paper presents SnipRec, a novel approach that uses snippets from user reviews and LLMs to enhance flexibility and reduce data annotation needs in CRS.

Findings

Outperforms document- and sentence-based methods in accuracy.

Effective across restaurant, book, and clothing domains.

Handles free-form user responses successfully.

Abstract

Conversational Recommender Systems (CRS) engage users in interactive dialogues to gather preferences and provide personalized recommendations. While existing studies have advanced conversational strategies, they often rely on predefined attributes or expensive, domain-specific annotated datasets, which limits their flexibility in handling diverse user preferences and adaptability across domains. We propose SnipRec, a novel resource-efficient approach that leverages user-generated content, such as customer reviews, to capture a broader range of user expressions. By employing large language models to map reviews and user responses into concise snippets, SnipRec represents user preferences and retrieves relevant items without the need for intensive manual data collection or fine-tuning. Experiments across the restaurant, book, and clothing domains show that snippet-based representations outperform document- and sentence-based representations, achieving Hits@10 of 0.25-0.55 with 3,000 to 10,000 candidate items while successfully handling free-form user responses.