Article Per Year (5 Year)
p-Index From 2021 - 2026
0.23
P-Index
This Author published in this journals
All Journal Journal of Computer Networks, Architecture and High Performance Computing
Marto Sihombing
Sekolah Tinggi Manajemen Informatika dan Komputer (STMIK) Kaputama, Binjai, Indonesia
Computer Science & IT Education

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

 1 
Implementation of K-Means Clustering for Grouping Post-Flood Disease Patterns in Affected Residential Settlements Hotler Manurung; Marto Sihombing; Ratih Puspadini
Journal of Computer Networks, Architecture and High Performance Computing Vol. 8 No. 3 (2026): Research Paper July 2026
Publisher : Information Technology and Science (ITScience)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47709/cnahpc.v8i3.9210

Abstract

Flooding in Binjai City increases post-flood disease incidence in slum settlements due to inadequate sanitation and high environmental vulnerability, while disease records are often scattered, poorly organized, and late, slowing health interventions. This study groups post-flood disease data in affected residential areas using K-Means clustering based on disease type, village, and slum level. A total of 1,100 records were collected from five districts in Binjai City; after excluding incomplete records, 850 valid records were used in the 2-cluster scenario and 1,086 in the 3-cluster scenario, implemented in a Matlab-based application. In the 2-cluster scenario, Cluster 1 contained 536 records with centroid (3.64, 12.48, 2.41) and Cluster 2 contained 314 records with centroid (4.18, 16.25, 2.09). In the 3-cluster scenario, Cluster 1 contained 498 records with centroid (3.45, 11.62, 2.31), Cluster 2 contained 312 records with centroid (4.02, 16.47, 2.05), and Cluster 3 contained 276 records with centroid (2.91, 8.35, 2.76), all dominated by diarrhea in medium-to-high slum-level areas. Internal validity indices (Silhouette, Davies-Bouldin, Calinski-Harabasz), computed on a reference sample, support retaining the 3-cluster scheme for its finer, more actionable risk stratification. The results show that K-Means clustering groups affected areas by disease and slum characteristics, and a centroid-derived priority ranking of the clusters is proposed to support health priority setting, medical resource distribution, and data-driven post-flood disease mitigation.
Co-Authors Hotler Manurung Ratih Puspadini