A Comprehensive Survey of Knowledge Graph Reasoning: Approaches and Applications
Guanglin Niu, Yangguang Lin, Bo Li
IEEE Transactions on Big Data
Problems Identified (5)
Limited KGR survey coverage: Previous KGR reviews cover only specific perspectives, leaving a need for a more comprehensive view of approaches and applications.
KGR deployment challenges: KGR faces open challenges involving trustworthiness, multimodal reasoning, continual learning, uncertainty, and LLM-driven approaches.
Theory-practice gap in KGR: There is a gap between theoretical advancements and practical deployment of KGR models.
Limited KGR survey coverage: Previous KGR reviews cover only specific perspectives, leaving a need for a more comprehensive view of approaches and applications.
KGR deployment challenges: KGR faces open challenges involving trustworthiness, multimodal reasoning, continual learning, uncertainty, and LLM-driven approaches.
Proposed Solutions (5)
Comprehensive KGR survey: The paper proposes a comprehensive survey of KGR covering foundational approaches and applications.
KGR approach review: The survey reviews seldom-attended and advanced KGR approaches including negative sampling, open-source libraries, rule-guided paradigms, and LLMs.
KGR application taxonomy: The paper provides a taxonomy of real-world KGR applications across horizontal and vertical domains.
Comprehensive KGR survey: The paper proposes a comprehensive survey of KGR covering foundational approaches and applications.
KGR approach review: The survey reviews seldom-attended and advanced KGR approaches including negative sampling, open-source libraries, rule-guided paradigms, and LLMs.
Results (3)
Comprehensive KGR perspective:
Model strengths and limitations analysis:
First real-world KGR application taxonomy:
Research Domain
Knowledge graph reasoning (KGR)