Application of Data Mining with K-Means Clustering for Disease Classification at the Pagambiran Health Center

Abstract

The Pagambiran Health Center is a place for public health services in Pagambiran Village, Padang City. Every day, the Pagambiran Health Center serves various community diseases. Many diseases and with different symptoms from the community make it difficult for the health center to determine the right counseling or socialization in a disease caused. Proper counseling will minimize the possibility of people contracting a disease. In addition, the health center still has difficulties in making decisions so that the results of counseling and socialization are not optimal. This study aims to produce a group of patient disease data and know the parameters of each patient's disease. The data used in this study is the patient's disease data that is in the patient's medical record data. In this study, the data used was patient disease data for a period of 5 months consisting of 268 patient disease categories. The data was processed using the K-Means method. The K-Means method  groups data with a partition system. The results obtained were patient diseases as many as 4 clusters, where cluster 1 was a disease that occurred with high intensity as many as 1 category of disease, cluster 2 was a disease that occurred with moderate intensity as many as 7 categories of diseases, cluster  3 was a disease that occurred with low intensity as many as 45 categories of diseases, and cluster 4 are diseases with very low/rare intensity as many as 215 disease categories. This study produces classification based on patient diseases so that the health center can carry out prevention and provide community services effectively because this study has produced models and considerations to determine the right treatment or countermeasure to the symptoms of community diseases.