OPTIMASI K-MEANS CLUSTERING UNTUK IDENTIFIKASI DAERAH ENDEMIK PENYAKIT MENULAR DENGAN ALGORITMA PARTICLE SWARM OPTIMIZATION DI KOTA SEMARANG

• Main Authors: Suhardi Rustam, Heru Agus Santoso, Catur Supriyanto
• Format: Article
• Language: English
• Published: Fakultas Ilmu Komputer UMI 2018-12-01
• Series: Ilkom Jurnal Ilmiah
• Subjects: endemic infectious disease; data mining; clustering; k-means; particle swarm optimization
• Online Access: http://jurnal.fikom.umi.ac.id/index.php/ILKOM/article/view/342

Tropical regions is a region endemic to various infectious diseases. At the same time an area of high potential for the presence of infectious diseases. Infectious diseases still a major public health problem in Indonesia. Identification of endemic areas of infectious diseases is an important issue in the field of health, the average level of patients with physical disabilities and death are sourced from infectious diseases. Data Mining in its development into one of the main trends in the processing of the data. Data Mining could effectively identify the endemic regions of hubunngan between variables. K-means algorithm klustering used to classify the endemic areas so that the identification of endemic infectious diseases can be achieved with the level of validation that the maximum in the clustering. The use of optimization to identify the endemic areas of infectious diseases combines k-means clustering algorithm with optimization particle swarm optimization ( PSO ). the results of the experiment are endemic to the k-means algorithm with iteration =10, the K-Fold =2 has Index davies bauldin = 0.169 and k-means algorithm with PSO, iteration = 10, the K-Fold = 5, index davies bouldin = 0.113. k-fold = 5 has better performance.