Summarizing online user reviews using bicliques
Paper in proceeding, 2016

With vast amounts of text being available in electronic format, such as news and social media, automatic multi-document summarization can help extract the most important information. We present and evaluate a novel method for automatic extractive multi-document summarization. The method is purely combinatorial, based on bicliques in the bipartite word-sentence occurrence graph. It is particularly suited for collections of very short, independently written texts (often single sentences) with many repeated phrases, such as customer reviews of products. The method can run in subquadratic time in the number of documents, which is relevant for the application to large collections of documents.

extractive summarization

bipartite clique

word frequency

Author

Muhammad Azam Sheikh

Saudi Electronic University (SEU)

Peter Damaschke

Chalmers, Computer Science and Engineering (Chalmers), Computing Science (Chalmers)

Olof Mogren

Chalmers, Computer Science and Engineering (Chalmers), Computing Science (Chalmers)

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

03029743 (ISSN) 16113349 (eISSN)

Vol. 9587 569-579
9783662491911 (ISBN)

Areas of Advance

Information and Communication Technology

Subject Categories

Computational Mathematics

Information Science

Roots

Basic sciences

DOI

10.1007/978-3-662-49192-8_46

ISBN

9783662491911

More information

Created

10/8/2017