Claim Missing Document
Check
Articles

Found 1 Documents
Search
Journal : JOURNAL OF SCIENCE AND SOCIAL RESEARCH

ANALISIS TINGKAT KEPARAHAN COVID-19 DI SUATU NEGARA MENGGUNAKAN METODE K-MEDOID CLUSTERING Amelia, Syahputri; Yusri, Eldo; Irwansyah, Bambang; Andira, Ayu; Sukri, Muhammad; Yafi, Muhammad Fauzan; Affandi, Muhammad
JOURNAL OF SCIENCE AND SOCIAL RESEARCH Vol 9, No 1 (2026): February 2026
Publisher : Smart Education

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.54314/jssr.v9i1.5787

Abstract

Abstract: This study aims to classify the severity of COVID-19 cases based on patient and region data using the K-Medoid Clustering method. COVID-19 has varying degrees of symptom severity, requiring cluster analysis to identify severity patterns to support decision-making in healthcare resource allocation and policy formulation. The data used included the number of positive cases, recovered cases, deaths, the average age of patients, and comorbidity levels. The results showed that the K-Medoid method was able to effectively cluster the data. In the raw dataset, the percentage of patients not infected with COVID-19 was 62.62%, while the percentage of infected patients was 37.38%. Based on sample characteristics, non-obese patients accounted for 74.54%, obese patients 25.46%, and patients with a combination of obesity and cardiovascular disease 0.57%. Keywords: Covid-19, Severity, K-Medoid Clustering, Data Mining Abstrak: Penelitian ini bertujuan untuk mengelompokkan tingkat keparahan kasus COVID-19 berdasarkan data pasien dan wilayah menggunakan metode K-Medoid Clustering. COVID-19 memiliki variasi tingkat keparahan gejala, sehingga diperlukan analisis klaster untuk mengidentifikasi pola keparahan yang mendukung pengambilan keputusan dalam alokasi sumber daya kesehatan dan perumusan kebijakan. Data yang digunakan meliputi jumlah kasus positif, kasus sembuh, kasus meninggal, usia rata-rata pasien, serta tingkat komorbiditas. Hasil penelitian menunjukkan bahwa metode K-Medoid mampu melakukan pengelompokan data secara efektif. Pada dataset mentah, persentase pasien tidak terjangkit COVID-19 sebesar 62,62%, sedangkan pasien terjangkit sebesar 37,38%. Berdasarkan karakteristik sampel, pasien non-obesitas memiliki persentase 74,54%, pasien obesitas 25,46%, dan pasien dengan kombinasi obesitas serta penyakit kardiovaskular sebesar 0,57%. Kata Kunci : Covid-19, Tingkat Keparahan, K-Medoid Clustering, Data Mining