<![CDATA[This study aims to analyze book borrowing patterns at the Library of the Faculty of Economics and Business, National University using the Apriori algorithm. The analysis was conducted on borrowing transaction data to identify relationships among book categories that are frequently borrowed together. The dataset consisted of 68 borrowing transactions covering 11 book categories. Although the number of transactions was relatively limited, the data were selected based on the availability of complete borrowing records during the observation period and were considered sufficient to identify initial borrowing patterns. The results reveal several significant patterns. The “Management” category obtained the highest support value of 40%, indicating that it was the most frequently borrowed category, while the rule “Management → General Management” achieved a confidence value of 70%, showing a strong tendency for both categories to be borrowed together. These findings demonstrate that the Apriori algorithm can effectively identify user borrowing preferences from circulation data. This study contributes to the development of data mining applications in library science, particularly in the use of association analysis to support evidence-based library management. The findings may assist librarians in optimizing collection arrangement, developing recommendation systems, and improving collection development strategies. Furthermore, this study highlights the potential of transaction data analysis as a practical approach for understanding user information needs in academic libraries.]]>
