Text2Bundle: Towards Personalized Query-based Bundle Generation

TL;DR

This paper introduces Text2Bundle, a novel framework for personalized, query-based bundle generation that combines user queries and historical preferences to generate tailored item bundles, outperforming existing methods.

Contribution

The paper proposes a new task of query-based bundle generation and introduces a framework that integrates query interests with long-term preferences using reinforcement learning.

Findings

Text2Bundle effectively leverages both query and historical data.

The framework outperforms state-of-the-art methods on real-world datasets.

Reinforcement learning enhances bundle relevance and personalization.

Abstract

Bundle generation aims to provide a bundle of items for the user, and has been widely studied and applied on online service platforms. Existing bundle generation methods mainly utilized user's preference from historical interactions in common recommendation paradigm, and ignored the potential textual query which is user's current explicit intention. There can be a scenario in which a user proactively queries a bundle with some natural language description, the system should be able to generate a bundle that exactly matches the user's intention through the user's query and preferences. In this work, we define this user-friendly scenario as Query-based Bundle Generation task and propose a novel framework Text2Bundle that leverages both the user's short-term interests from the query and the user's long-term preferences from the historical interactions. Our framework consists of three modules: (1) a query interest extractor that mines the user's fine-grained interests from the query; (2) a unified state encoder that learns the current bundle context state and the user's preferences based on historical interaction and current query; and (3) a bundle generator that generates personalized and complementary bundles using a reinforcement learning with specifically designed rewards. We conduct extensive experiments on three real-world datasets and demonstrate the effectiveness of our framework compared with several state-of-the-art methods.