Optimasi Nilai K Pada Algoritma k-Means untuk Klasterisasi Data Pasien Covid-19

Abstract

With the spread of Covid-19 to various countries, it is difficult for Governments and Health Agencies in the world to handle Covid-19 cases to date. The prevention carried out by the Government and Health Agencies in the world is carried out by giving vaccines to the public. However, in some places it is not implemented in accordance with PMK Number 84 of 2020 which prioritizes providing vaccines to the elderly. With the current density of the population in Indonesia, the administration of vaccines does not see who is prioritized first. The application of the k-means algorithm is carried out to cluster patients affected by Covid-19 on the Covid-19 case data obtained from kaggle.com in the form of patient data from January 1, 2020 to May 31, 2020 as many as 139119 cases. The results of clustering data on cases affected by Covid-19 with k=3 yielded a WCSS value of 6801292.2. Calculations of the K-Means Algorithm using the Google Collaboratory Tools resulted in clusters with the cases of patients affected by Covid-19 in Cluster-0 as many as 58.237 cases, in Cluster-1 as many as 53.932 cases, and in Cluster-2 as many as 26.950 cases.

Penerapan Machine Learning Clustering K-Means dan Linear Regression Dalam Penentuan Tingkat Resiko Tuberkulosis Paru

“…Oleh sebab itu, lebih baik dilakukan suatu proses inisialisasi cluster pemodelan Clusterwise Regression di tahapan seterusnya. Salah satu trik yang ada dalam inisialisasi cluster dengan metode Clustering [21], [22] Hasil pada tabel 5 merupakan penyelesaian proses iterasi centroid 3. Kemudian dilakukan proses iterasi ke 4 dengan menggunakan centroid centroid baru.…”

Mengelompokkan Daerah Rawan Kecelakaan Di Sumatera Utara dengan Algoritma Clustering

“…K-Means merupakan salah satu metode clustering data berbasis jarak dengan cara membagi data menjadi beberapa kelompok dengan tipe data numerik [6].…”