Development and application of an algorithm that extracts and learns customer needs by analyzing multilingual word-of-mouth information

2022 Fiscal Year Final Research Report

• Project/Area Number: 19K12231
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Review Section: Basic Section 62020:Web informatics and service informatics-related
• Research Institution: University of Tsukuba
• Principal Investigator: TSUDA Kazuhiko 筑波大学, ビジネスサイエンス系, 教授 (50302378)
• Co-Investigator(Kenkyū-buntansha): 吉田 健一 筑波大学, ビジネスサイエンス系, 教授 (40344858) ; 倉橋 節也 筑波大学, ビジネスサイエンス系, 教授 (40431663)
• Project Period (FY): 2019-04-01 – 2023-03-31
• Keywords: テキストマイニング / 自然言語処理 / 感情分析

Outline of Final Research Achievements

This research was conducted with the aim of extracting the true needs of customers in multiple countries by analyzing word-of-mouth information written in many languages on the Internet. However, because it was conducted during the COVID-19 pandemic, it was not possible to fully conduct a field survey on the positive and negative information that local people have regarding overseas word-of-mouth information.
However, regarding Thai and English, we obtained research results that suggest that it is possible to obtain positive and negative information that takes national characteristics into consideration. In addition, using the results obtained in this research, it was confirmed that there is a possibility that needs can be grasped by the proposed method by applying it to free description analysis of questionnaires in education and product analysis.The results were published in international papers.

Free Research Field

テキストマイニング

Academic Significance and Societal Importance of the Research Achievements

本研究の成果は，ただ単に記載内容を理解し，その文中から顧客ニーズを取得するだけでなく，その言語を母国語とする人たちの国民性を考慮して，記載文から感性情報やニーズを把握することにある．
特にタイ人における，人前では他人の悪口を記載しないという国民性から，文面の表層は褒めているポジティブな文書であっても，その裏を読むような思考ロジックがあることを確認し，この国民性を反映するニーズ抽出に取り組んだ．本アルゴリズムの精度が向上すれば，真のニーズが容易に取得できるので，社会的意義は大きいと言える．