Exploitation of Named Entities in Automatic Text Summarization for Swedish
2003 (English)In: Proceedings of NODALIDA’03 – 14th Nordic Conferenceon Computational Linguistics, Reykjavik, Iceland, May 30–31 2003gs of ., 2003, 9- p.Conference paper (Other academic)
Named Entities are often seen as important cues to the topic of a text. Theyare among the most information dense tokens of the text and largely definethe domain of the text. Therefore, Named Entity Recognition should greatlyenhance the identification of important text segments when used by an (extractionbased) automatic text summarizer. We have compared Gold Standardsummaries produced by majority votes over a number of manually createdextracts with extracts created with our extraction based summarization system,SweSum. Furthermore we have taken an in-depth look at how overweightingof named entities affects the resulting summary and come to theconclusion that weighting of named entities should be carefully consideredwhen used in a naïve fashion.
Place, publisher, year, edition, pages
2003. 9- p.
IdentifiersURN: urn:nbn:se:kth:diva-14081OAI: oai:DiVA.org:kth-14081DiVA: diva2:329599
NODALIDA’03 – 14th Nordic Conferenceon Computational Linguistics, Reykjavik, Iceland, May 30–31 2003
QC 201007122010-07-122010-07-122010-07-16Bibliographically approved