Toward Understanding the User Behavior in Sports University Library Using Hierarchical Clustering

Abstract

The library plays an important role in higher education. the emerging electronic media and digital contents bring libraries to encounter digital innovation and changes in leadership styles. However, analysis of the behavior of book borrowing is still a way to understand the demand of users, so as to provide sufficient resources and develop customized services. The purpose of this paper is to analyze the book borrowing data of the library in order to identify the user typologies. Numerous data records were collected from a sports university library in Taiwan. Each borrowing history record contains book detail and classification number. The characteristics of user behavior were described based on these data after cleaning, aggregation, and transforming. The hierarchical clustering techniques were applied to obtain the user typologies with similar behaviors. Five clusters, the general casual reader, athletes, Art and Literature lovers, course learner, and knowledge seeker, were obtained to represent the classic user typologies of a sports university.