Optimasi Nilai k Pada Algoritma K-Nearest Neighbor Untuk Klasifikasi Pasien Covid-19 Yang Membutuhkan Ruangan ICU

Abstract

Abstrack - The Coronavirus disease pandemic has caused many victims to fall due to being infected with the SARS-CO virus, so medical action is needed in hospitals to treat and stop the circulation of the viral infection. Various kinds of medical actions are carried out including treatment in the ICU (Intensive Care Unit). This is because those infected with COVID-19 can lead to more severe infections, they can develop organ failure or have a risk of death. This study aims to classify patients affected by COVID-19, especially those requiring treatment and care in the ICU. The results of this research can be utilized as a material for though for the hospital or other related parties to take policies in handling COVID-19 patients requiring treatment in the ICU. The classification will be carried out using the K-Nearest Neighbor algorithm and optimization on k value using k-fold cross validation 5-fold cross validation algorithm. The outcomes obtained are the value of k = 16, and the value of the performance accuracy test is 86.47%.