Denoising distant supervision for ontology lexicalization using semantic similarity measures

作者：

Highlights：

• Denoising the distant supervision assumption using semantic similarity is proposed.

• Different semantic similarity measures are proposed using word embeddings.

• The proposed method can be applied in any ontology lexicalization framework.

• Results show a significant increase in the generated ontology lexicon’s quality.

摘要

•Denoising the distant supervision assumption using semantic similarity is proposed.•Different semantic similarity measures are proposed using word embeddings.•The proposed method can be applied in any ontology lexicalization framework.•Results show a significant increase in the generated ontology lexicon’s quality.

论文关键词：Ontology lexicalization,Distant supervision,Denoising,Word embeddings

论文评审过程：Received 3 February 2020, Revised 11 March 2021, Accepted 16 March 2021, Available online 20 March 2021, Version of Record 10 April 2021.

论文官网地址：https://doi.org/10.1016/j.eswa.2021.114922