Diagnosa Penyakit Pneumonia Pada Anak Menggunakan Algoritma KNN

Abstract

Pneumonia is the leading cause of death in infants in the world, based on data from the Directorate General of P2PL Kemenkes RI in 2016 there are 503,738 the number of patients with pneumonia disease and the number of deaths of 551 infants. In giving initial treatment then created a technology such as aplication, which is expected to help parents to diagnose so that can be given initial treatment to the child. This study was made using K-Nearest Neighbor method that will diagnose pneumonia in children through clinical symptoms. This study was developed based on the results of interviews with doctors and supporting modules on pneumonia. Continued collection of data of patients with pneumonia used as training data .. This expert system is web-based ASP.NET and uses MariaDB as data storage. Based on test result for algorithm in application that using confusion matrix, the accuracy percentage is 89,10% from 156 data testing. Wuth that accuracy value K-Nearest Neighbor algorithm can give diagnose result for pneumonia, for giving parents to know the result and giving initial treatment to the child.