<![CDATA[With advances in information technology, libraries are experiencing significant digital transformation. In addition to the migration of physical books to digital format, the development of advanced library systems allows unrestricted access. However, with the diversity of books available, users cannot navigate the type of book to borrow next. This problem can be solved with a recommendation system approach using the Apriori algorithm. This algorithm helps in identifying patterns that appear regularly in data, such as relationships between items in a transaction database. Data is collected from lending transactions, and through a pre-processing process to produce consistent data, the system implements the Apriori algorithm with minimal variation in support values. and minimum confidence. The results of model evaluation based on the minimum support value (0.02 to 0.07) show a decrease in the number of frequent itemsets and rules along with an increase in the minimum support value. Meanwhile, testing is based on the minimum confidence value. An increase in the minimum confidence value from 0.1 to 0.5 causes a decrease in the number of rules and frequent itemsets, indicating increasingly strict association rules. However, when the confidence value decreases from 0.3 to 0.4, the number of itemsets and rules remains constant.]]>
Copyrights © 2024
