A Survey on Knowledge Graph for Enhancing the Performance of LLM-Based Recommendation Systems
Nurul Arina, Noor Akhmad Setiawan, Indriana Hidayah
Problems Identified (4)
KG-LLM recommender review gap: Recent reviews discuss KGs for cloud-based LLMs but rarely focus on their use in recommender systems.
Recommendation quality and explainability: The paper frames the need to improve recommendation quality, context, and explainability by combining LLMs and KGs.
KG-LLM recommender review gap: Recent reviews discuss KGs for cloud-based LLMs but rarely focus on their use in recommender systems.
Recommendation quality and explainability: The paper frames the need to improve recommendation quality, context, and explainability by combining LLMs and KGs.
Proposed Solutions (4)
KG-LLM recommender survey: The paper provides a literature review examining how knowledge graphs and large language models work together to enhance recommendations.
PRISMA paper analysis: The study uses the PRISMA approach to select and analyze 20 relevant papers on KG-LLM recommender systems.
KG-LLM recommender survey: The paper provides a literature review examining how knowledge graphs and large language models work together to enhance recommendations.
PRISMA paper analysis: The study uses the PRISMA approach to select and analyze 20 relevant papers on KG-LLM recommender systems.
Results (3)
Methods and metrics comparison:
Progress summary and future directions:
Methods and metrics comparison:
Research Domain
KG-LLM recommender systems