Experimental analysis of multiple criteria for extractive multi-document text summarization
作者：
Highlights：
• We focus on the different criteria used for generating automatic text summaries.
• We contribute with a complete study evaluating and comparing these different criteria.
• This paper analyzes the best criteria and their combinations.
• Our tests use Document Understanding Conferences datasets and the ROUGE metrics.
• Coverage, redundancy reduction, and relevance obtain the most balanced results.
摘要
•We focus on the different criteria used for generating automatic text summaries.•We contribute with a complete study evaluating and comparing these different criteria.•This paper analyzes the best criteria and their combinations.•Our tests use Document Understanding Conferences datasets and the ROUGE metrics.•Coverage, redundancy reduction, and relevance obtain the most balanced results.
论文关键词：Multi-document summarization,Multi-objective optimization,Content coverage,Redundancy reduction,Relevance,Coherence
论文评审过程：Received 16 January 2019, Revised 22 July 2019, Accepted 28 August 2019, Available online 29 August 2019, Version of Record 8 September 2019.
论文官网地址：https://doi.org/10.1016/j.eswa.2019.112904
