<![CDATA[The Online Public Access Catalog (OPAC) is a digital system that enables users to search for library references through an online interface using keywords. OPAC has been implemented to enhance IAIN Kediri library services. However, its usage has never been evaluated, resulting in limited understanding of user acceptance levels. This study aims to predict the acceptance of OPAC and identify the most influential variables using interpretable ensemble learning methods. This research used cross sectional design with data collected via a survey involving 400 IAIN Kediri students who had experience using the OPAC system. The study integrates the Technology Acceptance Model (TAM) with the Value-Based Adoption Model (VAM) framework. Predictor variables consist of Perceived Usefulness, Perceived Ease of Use, Intention, Technicality, and Enjoyment. The target variable was Actual Use. The measurement scale uses a Likert scale of 1 to 5. The instrument has been tested for validity and reliability. Ensemble learning algorithms used include Random Forest, AdaBoost, XGBoost, Lightgbm, and Catboost, with SHAP applied for model interpretability. Among the models tested, XGBoost achieved the highest predictive accuracy. SHAP analysis revealed that Enjoyment was the most significant factor influencing OPAC acceptance. These results demonstrate the effectiveness of interpretable ensemble models in predicting technology acceptance and suggest their potential as an alternative to data analysis methods. OPAC development can be done by improving the user interface and developing applications on Android.]]>
