Topic-Aware Knowledge Graph with Large Language Models for Interoperability in Recommender Systems

2024-12-28

AI Summary

To address cross-system semantic inconsistency, knowledge heterogeneity, and cold-start challenges in recommender systems, this paper proposes a topic-aware knowledge graph (KG) construction framework. Methodologically: (1) it introduces a novel hierarchical topic modeling mechanism that leverages large language models to iteratively extract general topics (capturing cross-domain commonalities) and specific topics (encoding fine-grained user preferences) from side information and contextual signals; (2) it designs synonym topic resolution and context-aware alignment algorithms to ensure semantic uniqueness and cross-system transferability; (3) it unifies topic representations via knowledge distillation and normalized refinement. Evaluated on multiple KG-based recommendation benchmarks, the framework achieves average improvements of 12.7% in Recall@20 and NDCG@20, significantly alleviating data sparsity and cold-start issues while enhancing cross-system interoperability.

Technology Category

• Data Mining & Knowledge Management: Recommender Systems
• Knowledge Representation and Reasoning: Knowledge Acquisition
• Machine Learning: Learning Preferences or Rankings

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs

Abstract

The use of knowledge graphs in recommender systems has become one of the common approaches to addressing data sparsity and cold start problems. Recent advances in large language models (LLMs) offer new possibilities for processing side and context information within knowledge graphs. However, consistent integration across various systems remains challenging due to the need for domain expert intervention and differences in system characteristics. To address these issues, we propose a consistent approach that extracts both general and specific topics from both side and context information using LLMs. First, general topics are iteratively extracted and updated from side information. Then, specific topics are extracted using context information. Finally, to address synonymous topics generated during the specific topic extraction process, a refining algorithm processes and resolves these issues effectively. This approach allows general topics to capture broad knowledge across diverse item characteristics, while specific topics emphasize detailed attributes, providing a more comprehensive understanding of the semantic features of items and the preferences of users. Experimental results demonstrate significant improvements in recommendation performance across diverse knowledge graphs.

Problem

• Integrating Large Language Models
• Thematic Knowledge Graphs
• Cross-system Collaboration

Innovation

• Large Language Models
• Knowledge Graph Integration
• Enhanced Recommendation Accuracy