Classification of Book Borrowing Interests Based on Class and Reading Category at the Mi Village Library of Tenggerejo using Naive Bayes

Abstract

School libraries play an essential role in supporting students’ literacy development. However, the low borrowing rate at MI Desa Tenggerejo indicates a mismatch between available collections and students’ reading interests. This study aimed to develop a classification system for book borrowing interest based on grade level and reading categories using the Naïve Bayes algorithm. The dataset included student information, reading categories, and borrowing history. The research process consisted of requirement analysis, system design, implementation, and model evaluation. Input variables included students’ grade level, reading categories, and book return punctuality, while the output variable was the borrowing interest level (High, Medium, Low). The implementation results showed that the system was successfully developed and could classify students’ borrowing interests according to actual data. The evaluation achieved an accuracy of 53.57%, with the highest misclassification occurring between the Medium and High categories. This indicated that although the system functioned as designed, the classification performance required further improvement through feature enrichment, balanced training data, and the application of alternative or ensemble algorithms. This study is expected to serve as a foundation for schools to design data-driven strategies to enhance literacy in primary education.