<![CDATA[This research aims to predicting the delay of book return in the Airlangga University library by using machine learning algorithms. With the consideration of approximately 1600 circulation documents from January to December 2023, several algorithms including Naïve Bayes, Support Vector Machine (SVM), Random Forest, Logistic Regression, Neural Networks, and Gradient Boosting are utilized. The Naïve Bayes model prove to be the most effective model by 92.7% accuracy and 97.7% precision in predicting return delay. The analysis of feature importance has demonstrated that a handful of features, especially days overdue, loan duration, and return date, are the main predictive variables for delay prediction in book returns outcomes. From this study, the Naïve Bayes can be an effective predictor of book return delays in Airlangga University library in order to improve user satisfaction, potentially notifying user in advance and offering alternatives. This study provided a promising picture about machine learning applications in library management systems for practical resource allocation and service quality improvement related to book return delays.]]>
Copyrights © 2025
