Presentation 2019-02-08
Selection of Figurative Expression Using Automatically Generated Data Set
Yuto Mitsuishi, Kazutaka Shimada,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Figurative expressions are one of effective approaches to lead readers to understand. In this paper, we propose a selection method of a suitable figurative expression about an adjective in asentence by using neural networks (LSTM with attention). NNs usually need a large amount of trainingdata for generating a model. We first construct the training date set automatically by computing a simile-likelihood score of each figurative candidate. Next, we select sentence-figurative expression pairs with high simile-likelihood scores as the training data. Then, we generate a LSTM-based selection model by using the training data. We obtained sentences with some interesting figurative expressions that cannot be selected by dictionary-based methods.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Automatically Generated Data Set / Automatic Selection / Deep Learning
Paper # NLC2018-42
Date of Issue 2019-01-31 (NLC)

Conference Information
Committee NLC / IPSJ-IFAT
Conference Date 2019/2/7(2days)
Place (in Japanese) (See Japanese page)
Place (in English) Ryukoku University Omiya Campus
Topics (in Japanese) (See Japanese page)
Topics (in English) The 14th Text Analytics Symposium
Chair Takeshi Sakaki(Hottolink)
Vice Chair Mitsuo Yoshida(Toyohashi Univ. of Tech.) / Kazutaka Shimada(Kyushu Inst. of Tech.)
Secretary Mitsuo Yoshida(Ryukoku Univ.) / Kazutaka Shimada(NTT)
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 Information Fundamentals and Access Technologies
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Selection of Figurative Expression Using Automatically Generated Data Set
Sub Title (in English)
Keyword(1) Automatically Generated Data Set
Keyword(2) Automatic Selection
Keyword(3) Deep Learning
1st Author's Name Yuto Mitsuishi
1st Author's Affiliation Kyushu Institute of Technology(Kyutech)
2nd Author's Name Kazutaka Shimada
2nd Author's Affiliation Kyushu Institute of Technology,(Kyutech)
Date 2019-02-08
Paper # NLC2018-42
Volume (vol) vol.118
Number (no) NLC-439
Page pp.pp.35-40(NLC),
#Pages 6
Date of Issue 2019-01-31 (NLC)