Presentation 2000/5/22
Automatic Extraction of Important Sentences and Its Valuation
Ren Fuji, Zhou Qiang, Shi Hongchi,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In this paper, we describe a method that uses both statistical and structural information in sentence extraction. In addition, following the analysis of human extractions, several heuristic rules are added to filter non-important sentences and to prevent similar sentence from being extracted. Our exprimental results proved the effectiveness of these means. In particular, once the heuristic rules being added, a significant improvement has been observed. How to evaluation of the quality of the extracted abstract is still an issue. Some methods have been presented to evaluate it. However, there is not a generally accepted good evaluation method. Based on the analysis of the distribution of human extracted important sentences we propose the evaluation criteria for the automatic important sentence extraction system. As the method uses double square roots, we call the method double square root method.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Natural Language Processing / Automatic Extraction of Important Sentence / Statistical Information / Structural Feature
Paper # NLC2000-3
Date of Issue

Conference Information
Committee NLC
Conference Date 2000/5/22(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Vice Chair

Paper Information
Registration To Natural Language Understanding and Models of Communication (NLC)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Automatic Extraction of Important Sentences and Its Valuation
Sub Title (in English)
Keyword(1) Natural Language Processing
Keyword(2) Automatic Extraction of Important Sentence
Keyword(3) Statistical Information
Keyword(4) Structural Feature
1st Author's Name Ren Fuji
1st Author's Affiliation Hiroshima City University()
2nd Author's Name Zhou Qiang
2nd Author's Affiliation Tsinghua University
3rd Author's Name Shi Hongchi
3rd Author's Affiliation University of Missouri
Date 2000/5/22
Paper # NLC2000-3
Volume (vol) vol.100
Number (no) 100
Page pp.pp.-
#Pages 7
Date of Issue