The impact of term-weighting schemes and similarity measures on extractive multi-document text summarization

作者：

Highlights：

• Extractive multi-document automatic text summarization is increasingly necessary.

• Its different term-weighting schemes and similarity measures are studied.

• All possible combinations of them (a total of 15) have been experimented and compared.

• As evaluation metrics, 8 different ROUGE scores and the execution time have been used.

• Term-frequency inverse-sentence-frequency and cosine similarity get the best results.

摘要

•Extractive multi-document automatic text summarization is increasingly necessary.•Its different term-weighting schemes and similarity measures are studied.•All possible combinations of them (a total of 15) have been experimented and compared.•As evaluation metrics, 8 different ROUGE scores and the execution time have been used.•Term-frequency inverse-sentence-frequency and cosine similarity get the best results.

论文关键词：Multi-document summarization,Extractive summary,Term-weighting,Similarity

论文评审过程：Received 16 January 2020, Revised 23 November 2020, Accepted 14 December 2020, Available online 24 December 2020, Version of Record 5 January 2021.

论文官网地址：https://doi.org/10.1016/j.eswa.2020.114510