IMPLEMENTASI K-MEANS UNTUK KLASTERISASI TINGKAT KEPARAHAN PENYAKIT PERIODONTAL BERDASARKAN DATA KLINIS

Abstract

Technological developments have had a significant impact in various fields, including health. In the field of dental health, one of the most common diseases is periodontal disease. Periodontal disease is a multifactorial disease that occurs in the supporting tissues of the teeth. The assessment of the severity of periodontal disease is often done subjectively. Therefore, a method is needed to help classify patient data based on the severity of periodontal disease in order to plan appropriate treatment based on the severity of the disease. This study uses the K-Means clustering algorithm. Clustering is performed based on gender, age, mean probing depth, mean clinical attachment loss, plaque index, and bleeding on probing. This study used the RapidMiner application and resulted in three levels of disease severity: mild, moderate, and severe. The evaluation results using the Davies-Bouldin Index showed a value of 0.688, indicating that the clusters formed had a good level of separation.