<![CDATA[Advances in health information technology require intelligent systems capable of supporting rapid and accurate diagnosis. This study proposes a Hybrid Recommender System (HRS) for preliminary medical diagnosis based on electronic medical records. The developed system combines K-Nearest Neighbor and Naïve Bayes Classifier with Multi-View TF-IDF feature representation. A total of 948 doctor-annotated medical records were used in the evaluation using a 10-Fold Cross-Validation scheme to improve the reliability of performance assessment. The results show that the hybrid model provides the best performance with an accuracy of 87.37% and an F1-score of 84.20%, consistently surpassing all comparison methods. These findings confirm that the integration of similarity-based and probabilistic learning can improve the quality of initial diagnosis recommendations in medical decision support systems. Further research will focus on expanding the dataset and clinical validation to ensure the reliability of the system in real-world practice.]]>
Copyrights © 2026
