Behavior Modeling Space Reconstruction for E-Commerce Search

2025-01-30

AI Summary

Existing e-commerce search behavior prediction models suffer from insufficient accuracy due to conflation of user preferences with query relevance and limited modeling capacity. To address this, we propose DRP—a Decoupled Preference modeling framework. First, we unify diagnostic analysis of modeling deficiencies using causal graphs and Venn diagrams. Second, we design a preference editing module to achieve interpretable decoupling of preference and relevance signals. Third, we introduce an adaptive fusion mechanism to dynamically coordinate their respective modeling pathways. DRP is the first framework to integrate causal inference with set-space analysis for search behavior modeling, effectively resolving the issue of modeling space collapse. Extensive experiments on two public benchmarks and an internal e-commerce dataset demonstrate consistent improvements: NDCG@10 increases by 3.2–5.8% over state-of-the-art methods, validating both efficacy and generalizability.

Technology Category

• Machine Learning: Learning Preferences or Rankings
• Knowledge Representation and Reasoning: Preferences
• Search and Optimization: Learning to Search

Application Category

• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Querying, indexing, and retrieval in Web-related graphs

Abstract

Delivering superior search services is crucial for enhancing customer experience and driving revenue growth. Conventionally, search systems model user behaviors by combining user preference and query item relevance statically, often through a fixed logical 'and' relationship. This paper reexamines existing approaches through a unified lens using both causal graphs and Venn diagrams, uncovering two prevalent yet significant issues: entangled preference and relevance effects, and a collapsed modeling space. To surmount these challenges, our research introduces a novel framework, DRP, which enhances search accuracy through two components to reconstruct the behavior modeling space. Specifically, we implement preference editing to proactively remove the relevance effect from preference predictions, yielding untainted user preferences. Additionally, we employ adaptive fusion, which dynamically adjusts fusion criteria to align with the varying patterns of relevance and preference, facilitating more nuanced and tailored behavior predictions within the reconstructed modeling space. Empirical validation on two public datasets and a proprietary search dataset underscores the superiority of our proposed methodology, demonstrating marked improvements in performance over existing approaches.

Problem

• Online Shopping
• Search Behavior Prediction
• Model Complexity

Innovation

• DRP method
• preference editing
• adaptive fusion