- 【Updated on May 12, 2025】 Integration of CiNii Dissertations and CiNii Books into CiNii Research
- Trial version of CiNii Research Automatic Translation feature is available on CiNii Labs
- Suspension and deletion of data provided by Nikkei BP
- Regarding the recording of “Research Data” and “Evidence Data”
Advantages and points to consider when applying quantitative text mining to qualitative Japanese text data: Based on a literature review of recent studies and linguistic features of Japanese words
Bibliographic Information
- Other Title
-
- 質的研究におけるテキストマイニング活用の利点と留意点―活用研究の検討と頻出単語の特徴をもとに―
Search this article
Description
本稿では,アンケートの自由記述やインタビューの発話などから得られたテキストデータを計量的に分析するテキストマイニングの活用を考察するため,異なる研究領域で発表された直近の論文を概観し,テキストマイニングを用いてどのようなテキストデータに対しどのような分析を行ったか,質的分析との併用はあったかを調べた.さらに論文の研究領域や用いたデータにかかわらず,頻出語に「思う」が出現することを指摘し,その理由を日本語学の知見から検討した.人間の書き言葉や話し言葉をデータとして用いるテキストマイニングにおいては,数字をデータにする計量的分析と異なり,日本語という言語や言語使用に対する知識と洞察に基づいた入力データの処理が不可欠であることを述べた.
In an attempt to explore advantages of conducting quantitative text mining to qualitative data, this study examines qualitative research papers published in 2018 from ten different research fields that have applied text-mining quantitative methods to Japanese qualitative text data. The review revealed quantitative text-mining was used effectively to help researchers obtain a general picture of the data and generate key categories. It also revealed that regardless of the types of data, one of the most frequently used words obtained by word-frequency analysis was “ think ” or to-omou in Japanese. This study draws on Japanese linguistics research to explain why “think” appears so frequently in the data. In Japanese, “ think ” or to-omou has a function similar to a modal verb that moderates a speakerʼs statement for an interpersonal purpose. This study suggests appropriately eliminating such words from the data before importing them for analysis.
Journal
-
- 札幌市立大学研究論文集 = SCU Journal of Design & Nursing
-
札幌市立大学研究論文集 = SCU Journal of Design & Nursing 13 (1), 47-53, 2019-07-30
札幌市立大学
- Tweet
Details 詳細情報について
-
- CRID
- 1390009224666995200
-
- NII Article ID
- 120006766192
-
- ISSN
- 18819427
-
- Web Site
- http://id.nii.ac.jp/1261/00000179/
-
- Text Lang
- ja
-
- Article Type
- departmental bulletin paper
-
- Data Source
-
- JaLC
- IRDB
- CiNii Articles
- KAKEN