Query expansion based on term distribution and DBpedia features

作者：

Highlights：

• Query expansion based on term distribution is better than Pseudo Relevance Feedback.

• The use of DBpedia to further term distribution expanded queries improves the recall.

• Query expansion is more sensitive to low numbers of expansion terms than to high ones.

• Mono-valued DBpedia attributes help in dealing with vocabulary mismatch.

摘要

•Query expansion based on term distribution is better than Pseudo Relevance Feedback.•The use of DBpedia to further term distribution expanded queries improves the recall.•Query expansion is more sensitive to low numbers of expansion terms than to high ones.•Mono-valued DBpedia attributes help in dealing with vocabulary mismatch.

论文关键词：Information retrieval,Query expansion,DBpedia spotlight,Term distribution,Language model

论文评审过程：Received 28 January 2020, Revised 5 August 2020, Accepted 9 March 2021, Available online 16 March 2021, Version of Record 26 March 2021.

论文官网地址：https://doi.org/10.1016/j.eswa.2021.114909