Double-hypergraph based sentence ranking for query-focused multi-document summarizaton

Xiaoyan Cai, Junwei Han, Lei Guo, Libin Yang

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

1 Scopus citations

Abstract

Traditional graph based sentence ranking approaches modeled the documents as a text graph where vertices represent sentences and edges represent pairwise similarity relationships between two sentences. Such modeling cannot capture complex group relationships shared among multiple sentences which can be useful for sentence ranking. In this paper, we propose two different group relationships (sentence-topic relationship and document-topic relationship) shared among sentences, and construct a double-hypergraph integrating these relationships into a unified framework. Then, a double-hypergraph based sentence ranking algorithm is developed for query-focused multi-document summarization, in which Markov random walk is defined on each hypergraph and the mixture Markov chains are formed so as to perform transductive learning in the double-hypergraph. When evaluated on DUC datasets, performance of the proposed approach is remarkable.

Original languageEnglish
Title of host publicationProceedings - 2016 IEEE/WIC/ACM International Conference on Web Intelligence Workshops, WIW 2016
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages112-118
Number of pages7
ISBN (Electronic)9781509047710
DOIs
StatePublished - 11 Jan 2017
Event2016 IEEE/WIC/ACM International Conference on Web Intelligence Workshops, WIW 2016 - Omaha, United States
Duration: 13 Oct 201616 Oct 2016

Publication series

NameProceedings - 2016 IEEE/WIC/ACM International Conference on Web Intelligence Workshops, WIW 2016

Conference

Conference2016 IEEE/WIC/ACM International Conference on Web Intelligence Workshops, WIW 2016
Country/TerritoryUnited States
CityOmaha
Period13/10/1616/10/16

Keywords

  • Hypergraph
  • Query-focused multi-document summarization
  • Sentence ranking

Fingerprint

Dive into the research topics of 'Double-hypergraph based sentence ranking for query-focused multi-document summarizaton'. Together they form a unique fingerprint.

Cite this