Presentation 2018-06-17
Effect of specifying the number of word occurrence in identification of specific text description
Yuki Okumura, Atsushi Moriyasu, Sachio Hirokawa, Kazuhiro Takeuchi,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Occurrence information of specific words is known as useful features for text classification. In this paper, we consider text classification using features that represent the number of occurrences in the text. Through such text classification, we aim to define more detailed feature to identify the texts that contain specific description of special concepts. For the purpose, we conduct an experiment using SVM to identify the texts that explain algorithms from the articles in Japanese Wikipedia. we analyze the results from the viewpoint of the writing style describing the specific concept.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) SVM / Sentence classification / term frequency information / attribute selection
Paper # TL2018-10
Date of Issue 2018-06-10 (TL)

Conference Information
Committee TL
Conference Date 2018/6/17(1days)
Place (in Japanese) (See Japanese page)
Place (in English) Kobe Gakuin University (Port Island Campus)
Topics (in Japanese) (See Japanese page)
Topics (in English) "Cognition in Passage of Time and their Expression","Language Education, Language and Education,"Language and Thought"
Chair Hiroshi Sano(Tokyo Univ. of Foreign Studies)
Vice Chair Tadahisa Kondo(Kogakuin Univ.) / Kazuhiro Takeuchi(Osaka Electro-Comm. Univ.)
Secretary Tadahisa Kondo(Kobe Gakuin Univ.) / Kazuhiro Takeuchi(Kyoto Inst. of Tech.)
Assistant Nobuyuki Jincho(Waseda Univ.) / Akinori Takada(Ferris Univ.) / Akio Ishikawa(KDDI Research)

Paper Information
Registration To Technical Committee on Thought and Language
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Effect of specifying the number of word occurrence in identification of specific text description
Sub Title (in English)
Keyword(1) SVM
Keyword(2) Sentence classification
Keyword(3) term frequency information
Keyword(4) attribute selection
1st Author's Name Yuki Okumura
1st Author's Affiliation Osaka Electro-Communication University(OECU)
2nd Author's Name Atsushi Moriyasu
2nd Author's Affiliation Osaka Electro-Communication University(OECU)
3rd Author's Name Sachio Hirokawa
3rd Author's Affiliation Kyusyu University(Kyusyu Univ.)
4th Author's Name Kazuhiro Takeuchi
4th Author's Affiliation Osaka Electro-Communication University(OECU)
Date 2018-06-17
Paper # TL2018-10
Volume (vol) vol.118
Number (no) TL-99
Page pp.pp.53-56(TL),
#Pages 4
Date of Issue 2018-06-10 (TL)