Entity summarization in fuzzy knowledge graph based on fuzzy concept analysis

Abstract

Considerable attention has recently been devoted to Knowledge Graph (KG), which has been applied in many domains. However, the information is often imprecise and vague when constructing the knowledge graph and thus the Fuzzy Knowledge Graph (FKG) emerged. Considering the increasing data in FKG, this paper firstly formulates the entity summarization in FKG and proposes an approach leveraging Fuzzy Formal Concept Analysis (FFCA). More specifically, the predicates and objects in RDF triples are deemed as attributes and objects in FFCA, respectively. Then, the fuzzy formal context can be obtained and the fuzzy concept lattice can be constructed. Finally, the concepts are ranked by the cardinality of the extent in concept lattice and the vague value of objects in RDF triples.