<![CDATA[The development of information technology in the field of data mining opens up significant opportunities to optimize data classification in the healthcare and rehabilitation sectors. LRPPN BI Medan currently faces challenges in determining rehabilitation programs because the decision-making process is still subjective and has not yet systematically utilized historical data. Inaccuracies in determining these programs can reduce recovery effectiveness and increase the potential for patient relapse. This study aims to apply and test the effectiveness of the KNN and Naïve Bayes algorithms in classifying rehabilitation programs based on patient criteria, such as duration of drug use, URICA test results, medical history, and addiction level. This study uses a quantitative approach with the Waterfall development method. This solution is proposed to overcome subjectivity through a data-based classification system that complements each other in accuracy and processing speed. The results of this study show that the Naive Bayes algorithm has a higher accuracy rate of 96.43%, while KNN has an accuracy of 94.29%.]]>
Copyrights © 2026
