テキストマイニングを用いたレシピ口コミ分析による味覚表現の意図解明

カテゴリ: 論文誌(論文単位)

グループ名: 【C】電子・情報・システム部門

発行日: 2021/10/01

タイトル(英語): A Method to Understand the Intention of Taste Expressions from Word-of-Mouth Data of Cooking Recipe Sites using Text Mining

著者名: 立花　伸一（筑波大学大学院），原田　智彦（筑波大学情報科学と社会科学の融合リサーチグループ），津田　和彦（筑波大学大学院）

著者名(英語): Shinichi Tachibana (University of Tsukuba), Tomohiko Harada (Research Group on Business Informatics on Social Science, University of Tsukuba), Kazuhiko Tsuda (University of Tsukuba)

キーワード: テキストマイニング，味覚表現，レシピ口コミサイト，BERT，LIME，共起分析 text mining，taste expression，word-of-mouth data of cooking recipe sites，BERT，LIME，co-occurrence analysis

要約(英語): The purpose of this study is to propose a quantitative evaluation method for clarifying intentions toward taste expressions that cannot be clarified, from word-of-mouth data of cooking recipe websites using text mining.This study aims to clarify the use of the word “KOKU” as an example to verify the method. As a result of applying the prediction results of BERT to LIME with certain changes, it was found that the characteristic ingredients contributing to the KOKU were identified, and that sweetness and umami were related to the KOKU as taste elements. As a result of the comparison of the terms used to describe the ingredients characteristic of KOKU and SAPPARI, there was a significant difference between the terms used to describe the oil and fat. Through the analysis mentioned above, the features of KOKU were defined. In the past, there has been no attempt to clarify the features of ambiguous taste expressions using word-of-mouth data from cooking recipe websites. The series of analysis methods presented in this paper enabled quantitative evaluation for clarifying the intentions of taste expressions that have been used by ordinary people without clear definitions.

本誌: 電気学会論文誌C（電子・情報・システム部門誌） Vol.141 No.10 （2021）

本誌掲載ページ: 1115-1124 p

原稿種別: 論文／日本語

電子版へのリンク: https://www.jstage.jst.go.jp/article/ieejeiss/141/10/141_1115/_article/-char/ja/