<![CDATA[In the digital era, the use of information technology in the health sector continues to grow, one of which is through the implementation of Electronic Medical Records (EMR). This study develops a web-based EMR system equipped with an automatic patient activity recommendation feature using the K-Nearest Neighbor (KNN) algorithm. Data from 550 patients from Dr. Viandini Clinic, Halo doc, and the internet were used with attributes of disease, age, blood pressure, and activity recommendations. The development process includes data collection, labeling, preprocessing, training and evaluation of the KNN model using the Accuracy@1 and Accuracy@5 metrics. The system is implemented with Laravel Filament and Python-Flask for the recommendation API. The test results show that the system is able to provide relevant recommendations with Accuracy@1 of 90.83% and Accuracy@5 of 95.31%. The application of KNN to this system supports automation, efficiency, and improvement of service quality in clinics and is the basis for the development of more personalized and data-driven digital health services.]]>
Copyrights © 2025
