Knowledge Graph Enhanced Language Agents for Recommendation

2024-10-25

arXiv.org

AI Summary

Existing recommender systems struggle to model deep semantic associations between users and items using language agents, resulting in coarse-grained user representations and weak recommendation interpretability. To address this, we propose KG-Path2Text—a novel framework that explicitly converts knowledge graph (KG) paths into natural language prompts and embeds them within a multi-agent collaborative simulation process. This enables language agents to dynamically construct fine-grained user profiles and attribute preferences via structured reasoning cues. Our key innovations include a KG-enhanced prompt injection mechanism and a path-to-text mapping strategy, jointly optimizing interpretable reasoning and dynamic profiling. Evaluated on three mainstream benchmarks, KG-Path2Text achieves 33%–95% improvements in NDCG@1 over state-of-the-art methods, significantly enhancing both recommendation accuracy and interpretability.

Technology Category

• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval
• Knowledge Representation and Reasoning: Preferences
• Planning, Routing, and Scheduling: Planning with Language Models

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Search and Retrieval-Augmented AI: Agentic search

Abstract

Language agents have recently been used to simulate human behavior and user-item interactions for recommendation systems. However, current language agent simulations do not understand the relationships between users and items, leading to inaccurate user profiles and ineffective recommendations. In this work, we explore the utility of Knowledge Graphs (KGs), which contain extensive and reliable relationships between users and items, for recommendation. Our key insight is that the paths in a KG can capture complex relationships between users and items, eliciting the underlying reasons for user preferences and enriching user profiles. Leveraging this insight, we propose Knowledge Graph Enhanced Language Agents(KGLA), a framework that unifies language agents and KG for recommendation systems. In the simulated recommendation scenario, we position the user and item within the KG and integrate KG paths as natural language descriptions into the simulation. This allows language agents to interact with each other and discover sufficient rationale behind their interactions, making the simulation more accurate and aligned with real-world cases, thus improving recommendation performance. Our experimental results show that KGLA significantly improves recommendation performance (with a 33%-95% boost in NDCG@1 among three widely used benchmarks) compared to the previous best baseline method.

Problem

• Recommendation Systems
• User-Item Deep Correlation
• Language Agent Understanding

Innovation

• KGLA
• Knowledge Graph
• Multi-hop Relationships