Ambiguous Query Answering with Neural Symbolic Reasoning Over Incomplete KG
Lihui Liu, Hanghang Tong
Synthesis lectures on computer science
Problems Identified (5)
Ambiguous KG query answering: The work addresses answering unclear or multi-meaning queries over knowledge graphs.
Query intent uncertainty: Ambiguous language creates uncertainty in interpreting the user's query intent.
Incomplete KG reasoning: The title frames the task as reasoning over incomplete knowledge graphs.
Ambiguous KG query answering: The work addresses answering unclear or multi-meaning queries over knowledge graphs.
Query intent uncertainty: Ambiguous language creates uncertainty in interpreting the user's query intent.
Proposed Solutions (4)
Neural-symbolic KG reasoning: The chapter applies neural-symbolic reasoning by combining symbolic knowledge-graph structure with neural network models to clarify ambiguous queries and retrieve relevant answers.
Semantic context query reformulation: The abstract describes using semantic analysis, context understanding, and query reformulation to address ambiguous queries.
Neural-symbolic KG reasoning: The chapter applies neural-symbolic reasoning by combining symbolic knowledge-graph structure with neural network models to clarify ambiguous queries and retrieve relevant answers.
Semantic context query reformulation: The abstract describes using semantic analysis, context understanding, and query reformulation to address ambiguous queries.
Results (3)
Ambiguity resolution:
Relevant accurate answers:
Ambiguity resolution:
Research Domain
Knowledge graph question answering / neural-symbolic reasoning