MAPS: Motivation-Aware Personalized Search via LLM-Driven Consultation Alignment

2025-03-03

AI Summary

Existing personalized search methods overlook users’ pre-search consultation behaviors, thus failing to capture their underlying intent. This paper addresses the e-commerce scenario and proposes the first dynamic modeling framework for the “consultation → intent evolution → search” process. We introduce a Mixture-of-Attention Experts (MoAE) mechanism coupled with a dual-alignment framework—integrating contrastive learning and bidirectional attention—to jointly model latent consultation-driven intent, query semantics, user preferences, and item features, while fusing heterogeneous textual sources (consultations, reviews, and product descriptions). Our approach effectively aligns cross-modal semantics, mitigates historical behavioral noise, and bridges category–text semantic gaps. Extensive experiments on both real-world and synthetic datasets demonstrate significant improvements: +12.6% in Recall@10 and +9.8% in NDCG@10 over state-of-the-art methods.

Technology Category

• Search and Optimization: Learning to Search
• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval
• Machine Learning: Mixture of Experts (MoE)

Application Category

• User Modeling, Personalization and Recommendation: ML for personalized search and recommendations
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Applications of semantic technologies for improving search, browsing, recommendation, personalization

Abstract

Personalized product search aims to retrieve and rank items that match users' preferences and search intent. Despite their effectiveness, existing approaches typically assume that users' query fully captures their real motivation. However, our analysis of a real-world e-commerce platform reveals that users often engage in relevant consultations before searching, indicating they refine intents through consultations based on motivation and need. The implied motivation in consultations is a key enhancing factor for personalized search. This unexplored area comes with new challenges including aligning contextual motivations with concise queries, bridging the category-text gap, and filtering noise within sequence history. To address these, we propose a Motivation-Aware Personalized Search (MAPS) method. It embeds queries and consultations into a unified semantic space via LLMs, utilizes a Mixture of Attention Experts (MoAE) to prioritize critical semantics, and introduces dual alignment: (1) contrastive learning aligns consultations, reviews, and product features; (2) bidirectional attention integrates motivation-aware embeddings with user preferences. Extensive experiments on real and synthetic data show MAPS outperforms existing methods in both retrieval and ranking tasks.

Problem

• Aligning contextual motivations with concise queries
• Bridging the category-text gap in personalized search
• Filtering noise within sequence history for better search

Innovation

• LLM-driven unified semantic space embedding
• Mixture of Attention Experts prioritizes semantics
• Dual alignment integrates motivation-aware embeddings