A Study on Information Retrieval for Large Document Corpora about Outcomes and Materials

2022 Fiscal Year Final Research Report

• Project/Area Number: 18K11986
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Review Section: Basic Section 90020:Library and information science, humanistic and social informatics-related
• Research Institution: Gunma University
• Principal Investigator: Yasukawa Michiko 群馬大学, 情報学部, 助教 (70361384)
• Project Period (FY): 2018-04-01 – 2023-03-31
• Keywords: 情報検索 / 情報組織化 / 情報資源の構築・管理 / 自然言語処理 / データベース

Outline of Final Research Achievements

This study focuses on the spelling mismatch of semantically similar index terms. This spelling mismatch is an important problem in the research field of information retrieval. Therefore, this study aimed to develop a new technology for extracting the semantic relationship between "outcomes (what is made)" and "materials (what is used)", which often inherently exists in large structured text data. In this study, we developed a method of feature analysis that can be applied to strategic database searches of research and educational information, which has been of growing interest in Institutional Research (IR) in recent years. We also applied the feature analysis method developed for the research field of IR to culinary information processing, and developed a method for efficiently retrieving the lists of ingredients and their related cooking methods of Japanese traditional cooking.

Free Research Field

情報学

Academic Significance and Societal Importance of the Research Achievements

本研究は、日本語や英語のテキストデータにおける文字の並びや単語の並びを利用する従来法では対処できなかった「成果物と材料の意味的関係」のデータ構造を効果的に捉えるための新たな検索技術を開発したという点に学術的意義がある。また、大学における研究と教育、および、家庭における調理と栄養管理は、国内外において普遍的な重要性を持つ主題であり、これらを検索対象とする情報検索に応用可能な特徴分析の手法を提案したという点において本研究の社会的意義は大きい。