A comparative study of positive and negative keywords based on text mining

Abstract

In writing a personal essay, it is crucial to put relevant keywords in the right places. Employing proper combinations of good keywords may leave a favorable impression on readers. On the other hand, misuse of keywords may result in a poor impression. This study analyzes keywords extracted from essays of passed and failed applicants to a management school. We apply text-mining functions provided by several R packages to examine critical factors related to the keywords. The frequency and structure of the keywords are investigated using association analysis techniques. According to the analysis, meaningful differences have been found in frequency, association and cluster of keywords between the group of essays that passed and those that failed. The methodology proposed in this study may be extended to other areas such as job interviews, marketing advertisements, an image makings. © 2015 ICIC International.