Litcius/Paper detail

Discourse-Aware Unsupervised Summarization for Long Scientific Documents

Yue Dong, A. Mircéa, Jackie Chi Kit Cheung

202143 citationsDOIOpen Access PDF

Abstract

We propose an unsupervised graph-based ranking model for extractive summarization of long scientific documents. Our method assumes a two-level hierarchical graph representation of the source document, and exploits asymmetrical positional cues to determine sentence importance. Results on the PubMed and arXiv datasets show that our approach 1 outperforms strong unsupervised baselines by wide margins in automatic metrics and human evaluation. In addition, it achieves performance comparable to many state-of-the-art supervised approaches which are trained on hundreds of thousands of examples. These results suggest that patterns in the discourse structure are a strong signal for determining importance in scientific articles.

Topics & Concepts

Automatic summarizationComputer scienceExploitSentenceGraphArtificial intelligenceRanking (information retrieval)Natural language processingInformation retrievalTopic modelRepresentation (politics)Machine learningTheoretical computer scienceLawPoliticsPolitical scienceComputer securityTopic ModelingAdvanced Text Analysis TechniquesBiomedical Text Mining and Ontologies
Discourse-Aware Unsupervised Summarization for Long Scientific Documents | Litcius