A Method of Extracting an Objective Index from Qualitative Data by Using Natural Language Processing : An Examination in the Vocational Rehabilitation.

DOI Web Site 2 Citations 14 References

Bibliographic Information

Other Title
  • 自然言語処理を利用して質的データから客観的評価指標を抽出する方法の検討 : 職業リハビリテーションにおける実証的検討
  • シゼン ゲンゴ ショリ オ リヨウ シテ シツテキ データ カラ キャッカンテキ ヒョウカ シヒョウ オ チュウシュツ スル ホウホウ ノ ケントウ ショクギョウ リハビリテーション ニ オケル ジッショウテキ ケントウ

Search this article

Abstract

The purpose of this study was to demonstrate a method of converting qualitative data into quantitative data to obtain an objective index. Evaluation of clients in educational and clinical settings is often written in free-format text. However, based on such qualitative data difficulty arises in judging common perspectives used by different evaluators. This paper presented a procedure of using qualitative data quantitatively to obtain an objective index from such data. Specifically, using morphological analysis and the tf-idf method, text data were converted into numerical data based on frequency of word appearance in the text on four domains in vocational rehabilitation (i. e., physical, mental, social, and vocational). The values were further analyzed with hierarchical cluster analysis. The analysis distinguished words appearing only in one of the domains (3 clusters) from those appearing across the domains. The study suggested a possibility that these words might be used as objective indices for client evaluation. The paper discusses issues such as necessity of developing exclusion rules of unsuitable words.

Journal

Citations (2)*help

See more

References(14)*help

See more

Keywords

Details 詳細情報について

Report a problem

Back to top