Extending latent semantic analysis to manage its syntactic blindness

作者：

Highlights：

• Latent Semantic Analysis is syntactically blind.

• LSA is unable to handle negation in sentences.

• xLSA uses Syntactic Information to enhance semantic similarity results.

• xLSA outperforms neural network-based models on simple inverse sentences.

摘要

•Latent Semantic Analysis is syntactically blind.•LSA is unable to handle negation in sentences.•xLSA uses Syntactic Information to enhance semantic similarity results.•xLSA outperforms neural network-based models on simple inverse sentences.

论文关键词：Natural Language Processing,Natural Language Understanding,Latent Semantic Analysis,Semantic Similarity

论文评审过程：Received 30 January 2020, Revised 26 July 2020, Accepted 14 October 2020, Available online 24 October 2020, Version of Record 11 December 2020.

论文官网地址：https://doi.org/10.1016/j.eswa.2020.114130