HTSS: A novel hybrid text summarisation and simplification architecture
作者：
Highlights：
• This paper presents a new hybrid loss function that enables the text summarisation model to generate easy-to-read simplified summaries.
• A new evaluation measure for the combined tasks of summarisation and simplification has been proposed.
• A novel parallel corpus of 5204 articles with their associated summarised simplified text for the combined task of text summasization and simplification has been provided for future research.
• The proposed hybrid approach outperforms existing state-of-the-art neural text simplification and abstractive text summarisation models by 38.94% and 53.40%, respectively.
摘要
•This paper presents a new hybrid loss function that enables the text summarisation model to generate easy-to-read simplified summaries.•A new evaluation measure for the combined tasks of summarisation and simplification has been proposed.•A novel parallel corpus of 5204 articles with their associated summarised simplified text for the combined task of text summasization and simplification has been provided for future research.•The proposed hybrid approach outperforms existing state-of-the-art neural text simplification and abstractive text summarisation models by 38.94% and 53.40%, respectively.
论文关键词：Summarisation,Simplification,Pointer generator,Deep learning
论文评审过程：Received 18 April 2020, Revised 21 May 2020, Accepted 24 June 2020, Available online 13 July 2020, Version of Record 13 July 2020.
论文官网地址：https://doi.org/10.1016/j.ipm.2020.102351
