Presentation 2018-06-01
Readability Categorization of Japan EIKEN Document using Machine Learning with TF-IDF
Rupasingha Arachchilage Hiruni Madhusha Rupasingha, Takeda Yui, Incheon Paik,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Understanding of the readability level and improvements of the text are needed for a specific audience. Accordingly, automatic measurement of text readability has been important issue and there have been many approaches to solve it. Classical readability is measured by readability formulas, and more recent research have employed machine learning algorithms. However, those machine learning approaches use complex features such as linguistic and grammatical features to hinder calculating correct readability score. In this paper, we investigate which features available as input to machine learning improves the performance more easily and accurately. We experiment the readability categorization by three kinds of feature vectors: readability score, Term Frequency-Inverse Document Frequency (TF-IDF), and the combination of them. The classification results by TF-IDF only give accurate results than other two features.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Readability / Eiken Document / Document Classification / Machine Learning
Paper # SC2018-7
Date of Issue 2018-05-25 (SC)

Conference Information
Committee SC
Conference Date 2018/6/1(2days)
Place (in Japanese) (See Japanese page)
Place (in English) UBIC 3D Theater, University of Aizu
Topics (in Japanese) (See Japanese page)
Topics (in English) Service Computing for the 4th Industrial Revolution and Other Issues
Chair Masahide Nakamura(Kobe Univ.)
Vice Chair Shinji Kikuchi(Univ. of Aizu) / Yoji Yamato(NTT)
Secretary Shinji Kikuchi(NEC) / Yoji Yamato(Fujitsu Lab.)
Assistant

Paper Information
Registration To Technical Committee on Service Computing
Language ENG
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Readability Categorization of Japan EIKEN Document using Machine Learning with TF-IDF
Sub Title (in English)
Keyword(1) Readability
Keyword(2) Eiken Document
Keyword(3) Document Classification
Keyword(4) Machine Learning
1st Author's Name Rupasingha Arachchilage Hiruni Madhusha Rupasingha
1st Author's Affiliation University of Aizu(UoA)
2nd Author's Name Takeda Yui
2nd Author's Affiliation University of Aizu(UoA)
3rd Author's Name Incheon Paik
3rd Author's Affiliation University of Aizu(UoA)
Date 2018-06-01
Paper # SC2018-7
Volume (vol) vol.118
Number (no) SC-72
Page pp.pp.37-42(SC),
#Pages 6
Date of Issue 2018-05-25 (SC)