Presentation 2013-03-07
Extraction of Similar Words Based on Adaptation and Time-correlation of Maximal Substrings from Tweets of The Same Topic
Yuichiro HISANO, Kazuhito SAWASE, Hajime NOBUHARA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In order to merge various onomastic expressions for valuable tweet topic retrieval/clustering, a construe- tion method of twitter dictionaries based on lexical extraction and their time-correlation is proposed. In this paper, we propose Maximal substrings to extract efficient lexical set and adaptation to remove superfluous substrings. Moreover, similarities between keywords are calculated by the time-correlation of each word and co-occurrence probability. Through experiments with 101,714/354,706 tweets with the hashtags related to "NHK Kohaku-Uta- gassen "in 2011/2012, the effectiveness of the proposed method compared with the method used morphological analysis is shown.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Microblog / Retrieval support / Time-correlation / Maximal substring / Adaptation
Paper # SIS2012-49
Date of Issue

Conference Information
Committee SIS
Conference Date 2013/2/28(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 Smart Info-Media Systems (SIS)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Extraction of Similar Words Based on Adaptation and Time-correlation of Maximal Substrings from Tweets of The Same Topic
Sub Title (in English)
Keyword(1) Microblog
Keyword(2) Retrieval support
Keyword(3) Time-correlation
Keyword(4) Maximal substring
Keyword(5) Adaptation
1st Author's Name Yuichiro HISANO
1st Author's Affiliation University of Tsukuba()
2nd Author's Name Kazuhito SAWASE
2nd Author's Affiliation University of Tsukuba
3rd Author's Name Hajime NOBUHARA
3rd Author's Affiliation University of Tsukuba
Date 2013-03-07
Paper # SIS2012-49
Volume (vol) vol.112
Number (no) 465
Page pp.pp.-
#Pages 6
Date of Issue