Klustering Penyakit Tuberkulosis Menggunakan Algoritma K-Means (Studi Kasus Puskesmas Cisarua)

Abstrak

Tuberculosis (TB) is an infectious infection caused by the bacteria Mycobacterium tuberculosis, which generally attacks the lungs and can affect other organs. Although treatment with antibiotics is available, challenges such as drug resistance and lack of access to health services still hamper TB control efforts. TB is very affected in the environment, so we carry out screening in each area to find out whether or not many residents are affected by TB disease. Where in the Cisarua Health Center there are several areas affected by TB disease. TB disease can also be a benchmark for how good the level of health services in an area is because TB is also included in 

the government program. Referring to the existing phenomenon, the author was encouraged to carry out clustering of TB disease data at the Cisarua Health Center. The method used to group the data is the k-means algorithm. Referring to the test findings, it was found that the optimal k value is k=3. Thus, to group the dataset used into 3 cluster groups, where cluster_0 has 245 data, cluster_1 has 397 data and cluster_2 has 250 data. Where in cluster 1 it is dominated by RW 11 with the predominance of the disease being dominated by TB disease and mostly male patients suffering from it, the distribution based on age is dominated by the adult category (19-59). Meanwhile, in cluster_0, many patients were treated for tuberculosis, data dominated by RW 14, which mostly attacked female and male patients. Meanwhile, the distribution based on age categories is dominated by the adult category. The k-means algorithm obtained a Davies Bouldin value of -0.906. The findings of this study are projected to be able to be used as a reference source for the Cisarua Community Health Center to implement health promotion in its area.