EdgeSumm: Graph-based framework for automatic text summarization

作者：

Highlights：

• Proposing a graph-based framework for extractive single-document summarization.

• Combining a set of extractive methods to benefit from their advantages.

• Proposing a new text graph model and adjusting the graph nodes weights dynamically.

• Evaluating the proposed framework using ROUGE on the DUC2001 and DUC2002 datasets.

• The evaluation results show that Edgesumm outperforms the state-of-the-art ATS systems.

摘要

•Proposing a graph-based framework for extractive single-document summarization.•Combining a set of extractive methods to benefit from their advantages.•Proposing a new text graph model and adjusting the graph nodes weights dynamically.•Evaluating the proposed framework using ROUGE on the DUC2001 and DUC2002 datasets.•The evaluation results show that Edgesumm outperforms the state-of-the-art ATS systems.

论文关键词：Automatic text summarization,Extractive text summarization,Graph representation model,Single-document summarization,EdgeSumm

论文评审过程：Received 21 January 2020, Revised 4 February 2020, Accepted 7 April 2020, Available online 30 June 2020, Version of Record 30 June 2020.

论文官网地址：https://doi.org/10.1016/j.ipm.2020.102264