A Novel Implementation of Library Personalized Recommendation Service based on Associative Rule Mining Model

Modern libraries are evolving into user-centered spaces, prioritizing customized services to meet diverse needs for information in the digital era. Understanding user preferences based on past usage patterns is essential for developing intelligent recommendation systems that guide individuals toward relevant resources. This study introduces a personalized recommendation model grounded in the suggested Apriori algorithm, which identifies frequent item co-occurrences and derives rules that best reflect user interests. The proposed algorithm examines historical transaction records to find co-occurring items and constructs association rules based on their statistical significance. Experiments were conducted using the publicly available Book-Crossing dataset, which encompasses a wide range of subject areas and user interactions. Key performance metrics, including Support (up to 18.6%), Confidence (up to 74.2%), and Lift values greater than 1.8, validate the effectiveness of the generated rules in capturing meaningful relationships among resources. The results indicate that the system delivers accurate and interpretable recommendations without relying on extensive computational resources. This rule based approach offers a lightweight and efficient solution for enhancing personalized library services. By providing contextually relevant suggestions, the framework promotes user engagement, supports efficient information discovery, and encourages access to underutilized materials. Overall, the study highlights the efficiency of association rule mining in developing scalable, data-driven recommendation systems tailored to the evolving demands of modern digital library environments.