Investigating factors affecting library visits by university students using data mining

Abstract

Background. Providing appropriate library services to students is a challenging task for university librarians. The library at Loei Rajabhat University has some concerns about its small number of visitors. The question of “what is known about the situation?” was raised. As an attempt to answer this question, data mining was employed to gain insights into library and student data. Objective. This study used two data mining algorithms—Naïve Bayes and C4.5 decision tree induction—to analyze the data. The results of the data mining were intended to be used in promoting undergraduate students to physically visit the library. Method. Data include students’ library gate entry collected from the library database and student data collected from the university registrar’s office. Results. The data mining yielded interesting results. Senior students were found to use the library less than younger students. There were two faculties whose students come to the library less than 50%. Current GPA was found to be an influential attribute for predicting library visit. Contributions. The research identified useful student attributes for predicting library visit. The results of the data mining can be used to increase the rate of library use by organizing activities that target those attributes. For example, the library can collaborate with the instructors to organize programs for students with low GPA.