A novel method based on symbolic regression for interpretable semantic similarity measurement

作者：

Highlights：

• First attempt to use symbolic regression in the domain of semantic similarity.

• Strategy capable of handling a trade-off between accuracy and interpretability.

• Resulting models are the most interpretable to date in the field of semantic similarity.

摘要

•First attempt to use symbolic regression in the domain of semantic similarity.•Strategy capable of handling a trade-off between accuracy and interpretability.•Resulting models are the most interpretable to date in the field of semantic similarity.

论文关键词：Knowledge engineering,Symbolic regression,Similarity learning,Semantic similarity measurement

论文评审过程：Received 9 December 2019, Revised 14 May 2020, Accepted 12 June 2020, Available online 26 June 2020, Version of Record 3 July 2020.

论文官网地址：https://doi.org/10.1016/j.eswa.2020.113663