SemPathFinder: Semantic path analysis for discovering publicly unknown knowledge

作者：

Highlights：

• The present paper proposes a new LBD system, called SemPathFinder, which provides semantic path analysis that enables storytelling for plausible hypothesis generation.

• The paper adopts advanced text mining techniques such as named entity recognition and relation extraction to create a knowledge graph.

• The paper utilizes UMLS to improve accuracy of extraction and obtains semantics of extracted entities and relations.

• The paper explores several different ranking approaches including semantic related scores for spotting plausible new hypotheses.

摘要

•The present paper proposes a new LBD system, called SemPathFinder, which provides semantic path analysis that enables storytelling for plausible hypothesis generation.•The paper adopts advanced text mining techniques such as named entity recognition and relation extraction to create a knowledge graph.•The paper utilizes UMLS to improve accuracy of extraction and obtains semantics of extracted entities and relations.•The paper explores several different ranking approaches including semantic related scores for spotting plausible new hypotheses.

论文关键词：Literature based discovery,Named entity recognition,Relation extraction,Semantic path analysis,Semantic relatedness score

论文评审过程：Received 16 March 2015, Revised 22 June 2015, Accepted 22 June 2015, Available online 1 September 2015, Version of Record 1 September 2015.

论文官网地址：https://doi.org/10.1016/j.joi.2015.06.004