Background Due to the recent closure of library spaces caused by COVID-19 and the emergence and development of various contactless book borrowing services, physical libraries have been partially or entirely replaced by digital libraries. To enhance ...
Background Due to the recent closure of library spaces caused by COVID-19 and the emergence and development of various contactless book borrowing services, physical libraries have been partially or entirely replaced by digital libraries. To enhance the user experience in searching for novels, it is essential to understand the emotional characteristics, thematic features, and user preferences of novels. Therefore, the researcher recognizes the need for sentiment analysis related to the book category of novel. This will provide fundamental information for designing visualizations in the search interface of digital library novels.
Methods The research methodology of this study involved a literature review followed by empirical analysis. Through literature review, theories on digital libraries, novels, and sentiment analysis were examined. In empirical analysis, KJ method, cluster analysis, survey, and principal component analysis were employed.
Result First, the reliability and usability of the novels showed a direct correlation. Second, the reliability of the novels showed an inverse correlation with its originality. Third, the usability of the novels showed an inverse correlation with its originality.
Conclusion Through sentiment analysis of the novels, we can observe the emotional characteristics, thematic features, and users' sentiment evaluations of novels within each category. The results of this study are expected to serve as foundational data for designing visualizations in the search interface of digital library novels. Based on this, we anticipate that improving the efficiency and accuracy of novels searches can be achieved through visual design in the search interface of the novels.