User Preference-Based Movie Recommender System Using Decision Tree Method

Yutika Amelia Effendi
Reizo Nararya Alinandito (Universitas Airlangga)
Muhammad Akmal Rinaldy (Universitas Airlangga)
Rizka Dwi Nurwicaksanti (Universitas Airlangga)
Muhammad Daffa Tristan (Universitas Airlangga)

Journal of Advanced Technology and Multidiscipline (JATM)
Vol. 5 No. 1 (2026): Journal of Advanced Technology and Multidiscipline

This study presents the design and implementation of a movie recommendation system aimed at improving content discovery amidst the rapid growth of the film industry. The system utilizes the Decision Tree Algorithm, a non-parametric supervised learning technique known for its hierarchical structure and interpretability, to provide personalized movie suggestions. Empirical evaluation demonstrates the Decision Tree's effectiveness, achieving a Mean Absolute Error (MAE) of 0.942, which outperforms a Collaborative Filtering baseline (MAE 1.242). This research highlights the practical utility of Decision Trees in enhancing recommendation accuracy and efficiency, contributing to transparent and effective AI systems for personalized content discovery.