Presentation 2018-06-02
Taxonomy classification with clustered ESA and Machine learning
Ryo Ataka, Incheon Paik,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) It is known that the classification of documents with taxonomically short distance has lower accuracy than that of usual documents. In this paper, we use Explicit Semantic Analysis (ESA) to extract additional information from input documents and improve taxonomy classification performance. ESA is used to create a semantic interpreter for semantic analysis. We propose a method to create a meaningful additional feature vector element with clustered ESA and combine it with the original document. Then we classify it by several machine learning algorithms. We gained high accuracy improvement by text classification using Convolutional Neural Networks (CNN) and Multilayer Perceptron (MLP).
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Taxonomy classification / Machine learning / ESA
Paper # SC2018-12
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.)

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) Taxonomy classification with clustered ESA and Machine learning
Sub Title (in English)
Keyword(1) Taxonomy classification
Keyword(2) Machine learning
Keyword(3) ESA
1st Author's Name Ryo Ataka
1st Author's Affiliation University of Aizu(UoA)
2nd Author's Name Incheon Paik
2nd Author's Affiliation University of Aizu(UoA)
Date 2018-06-02
Paper # SC2018-12
Volume (vol) vol.118
Number (no) SC-72
Page pp.pp.65-70(SC),
#Pages 6
Date of Issue 2018-05-25 (SC)