<![CDATA[This study analyzes book borrowing transaction data from the Jambi University Library to identify borrowing patterns and extract valuable insights. By utilizing the FP-Growth algorithm within the framework of association rules, the research aims to uncover frequent itemset patterns that reveal relationships between different categories of borrowed books. These patterns are crucial for supporting librarians in making informed decisions for effective library management. The dataset consists of 2,978 book borrowing transactions recorded in 2022. Using Python for computational analysis, the study identified 14 association rules by applying a minimum support threshold of 0.005 and a minimum confidence threshold of 0.1. The resulting association rules include the following pairs: Management and Economics (0.006), Agriculture and Economics (0.014), Psychology and Education (0.013), General Works and Education (0.026), Mathematics and Education (0.005), Mathematics and Science (0.006), Mathematics and Economics (0.006), Social Sciences and Law (0.007), Politics and Law (0.012), Politics and Social Sciences (0.005), and Fiction and Language (0.005). These association rules offer valuable insights that can assist librarians in optimizing the organization of book collections, prioritizing acquisitions, and making strategic decisions to enhance the quality of library services. This approach highlights the potential of data-driven decision-making to improve library operations and increase user satisfaction.]]>
