Presentation 2019-09-28
On evaluation of word-embedding models for technical term in space systems
Naoko Okubo, Sho Kurahayashi, Kenji Mori, Kohsuke Namihira, Yasushi Ueda, Masafumi Katahira, Toshiyuki Amagasa,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In this paper, we report the examination result of the method to evaluate the quality of the word-em-bedding models with word2vec before applying them to the actual NLP task. The quality about word-embeddingmodels can be confirmed when applied to a real NLP task. However, when the word-embedding model is applied to tasks that require specialized knowledge for the evaluation, such as spacecraft development, it is difficult to applymultiple evaluation costs because the number of people who can be evaluated is limited. Therefore, we examined a method to evaluate the quality of the word-embedding models with a certain degree of accuracy before applying it to a real NLP task. We also investigated the correlation between the evaluation result of the word-embedding model and the accuracy of the actual NLP task.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) word-embedding model / word2vec / technical term
Paper # NLC2019-28
Date of Issue 2019-09-20 (NLC)

Conference Information
Committee NLC / IPSJ-DC
Conference Date 2019/9/27(2days)
Place (in Japanese) (See Japanese page)
Place (in English) Future Corporation
Topics (in Japanese) (See Japanese page)
Topics (in English) The Thirteenth Text Analytics Symposium
Chair Takeshi Sakaki(Hottolink) / Ryoji Akimoto(Toppan Printing)
Vice Chair Mitsuo Yoshida(Toyohashi Univ. of Tech.) / Kazutaka Shimada(Kyushu Inst. of Tech.)
Secretary Mitsuo Yoshida(Ryukoku Univ.) / Kazutaka Shimada(NTT) / (Future Univ. Hakodate)
Assistant Takeshi Kobayakawa(NHK) / Hiroki Sakaji(Univ. of Tokyo)

Paper Information
Registration To Technical Committee on Natural Language Understanding and Models of Communication / Special Interest Group on Document Communication
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) On evaluation of word-embedding models for technical term in space systems
Sub Title (in English)
Keyword(1) word-embedding model
Keyword(2) word2vec
Keyword(3) technical term
1st Author's Name Naoko Okubo
1st Author's Affiliation Japan Aerospace Exploration Agency(JAXA)
2nd Author's Name Sho Kurahayashi
2nd Author's Affiliation Japan Aerospace Exploration Agency(JAXA)
3rd Author's Name Kenji Mori
3rd Author's Affiliation Japan Aerospace Exploration Agency(JAXA)
4th Author's Name Kohsuke Namihira
4th Author's Affiliation Japan Aerospace Exploration Agency(JAXA)
5th Author's Name Yasushi Ueda
5th Author's Affiliation Japan Aerospace Exploration Agency(JAXA)
6th Author's Name Masafumi Katahira
6th Author's Affiliation Japan Aerospace Exploration Agency(JAXA)
7th Author's Name Toshiyuki Amagasa
7th Author's Affiliation University of Tsukuba(Unv. Tsukuba)
Date 2019-09-28
Paper # NLC2019-28
Volume (vol) vol.119
Number (no) NLC-212
Page pp.pp.109-114(NLC),
#Pages 6
Date of Issue 2019-09-20 (NLC)