Perbandingan Metode Clustering dalam Pengelompokan Data Puskesmas pada Cakupan Imunisasi Dasar Lengkap

• Main Authors: Pelsri Ramadar Noor Saputra, Ahmad Chusyairi
• Format: Article
• Language: Indonesian
• Published: Ikatan Ahli Indormatika Indonesia 2020-12-01
• Series: Jurnal RESTI (Rekayasa Sistem dan Teknologi Informasi)
• Subjects: clustering; fuzzy c-means; k-means; puskesmas; silhouette coefficient
• Online Access: http://jurnal.iaii.or.id/index.php/RESTI/article/view/2556

The coverage of Health Care Center toward UCI (Universal Child Immunization) at Banyuwangi Regency in 2018 met the target 91%. Onfortunately with a high amount of immunization, the number of infant deaths reached 138 infants. Total number increased 111 from the previous year. A review of the complete basic immunization data needs to be done. In this research, a clustering method was proposed by comparing the K-Means and Fuzzy C-Means (FCM) algorithm in grouping Health Care Center data. Silhouette Coefficient and Standart Deviation were used to evaluate clusters that were perfomed to find out the accuracy in grouping data. The result showed that the FCM algorithm was better than K-Means based on Silhouette Coefficient results that were close to good, and the calculation of Standart Deviation had a smaller result that was 0.0918 than K-Means with the results of 0.0942. The Grouping of Heath Care Center data can be considered by the Health Department of Banyuwangi Regency in evaluating complete basic immunization services, especially in groups with poor immunization services to reduce infant and child mortality, so a disease that can be prevented with immunization become lower.