Presentation 2012-02-03
Complaint sentence detection via automatic training data generation using sentiment lexicons and context coherence
Yusuke UMEZAWA, Takashi INUI, Mikio YAMAMOTO,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) By the spread of the web in recent years, the reviews about products and services exist in large quantities on the web. If we can detect complaints from such reviews, they are valuable as information for users to prevent disadvantage and for companies to improve products. In this study, using sentiment lexicons and context coherence, we propose a method to generate training data tagged on from a large amount of reviews in large quantities on the web. Experiments of complaint sentence detection using a classifier trained the data. As a result of experiments, it was shown that the classifier is higher performance than using handmade but small training data.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) sentiment lexicons / context coherence / automatic training data generation / complant
Paper # NLC2011-64
Date of Issue

Conference Information
Committee NLC
Conference Date 2012/1/26(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

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) Complaint sentence detection via automatic training data generation using sentiment lexicons and context coherence
Sub Title (in English)
Keyword(1) sentiment lexicons
Keyword(2) context coherence
Keyword(3) automatic training data generation
Keyword(4) complant
1st Author's Name Yusuke UMEZAWA
1st Author's Affiliation Graduate School of Systems and Information Engineering, University of Tsukuba()
2nd Author's Name Takashi INUI
2nd Author's Affiliation Systems and Information, University of Tsukuba
3rd Author's Name Mikio YAMAMOTO
3rd Author's Affiliation Systems and Information, University of Tsukuba
Date 2012-02-03
Paper # NLC2011-64
Volume (vol) vol.111
Number (no) 427
Page pp.pp.-
#Pages 6
Date of Issue