Wihendra, Tomi (2025) Perbandingan Metode Naive Bayes dan K-nearest Neighbor dalam Klasifikasi Penyakit Ispa (Studi Kasus : Rumah Sakit Permata Hati Kota Duri). Other thesis, Universitas Islam Riau.

Text skripsi_203510398_watermark.pdf - Submitted Version Restricted to Registered users only Download (5MB) | Request a copy

Abstract

Acute Respiratory Infection (ARI) is a common and highly contagious illness, especially affecting children and the elderly. The diagnostic process at Permata Hati Hospital in Duri City is still conducted manually, which is time-consuming and highly dependent on medical expertise. This study aims to compare two machine learning classification methods, namely Naïve Bayes and K-Nearest Neighbor (KNN), in classifying types of ARI based on patient data. The dataset used consists of secondary data from ARI patients collected between 2022 and 2023, including symptoms and diagnoses. A web-based classification system was developed using Python programming language and Flask framework. The performance of both algorithms was evaluated using metrics such as accuracy, precision, recall, and F1-score. Model testing using a confusion matrix in the classification of ISPA disease with 233 training data and 59 test data shows that the KNN method has better classification performance with an accuracy level of of 96.6%, precision of 94.1%, recall of 91.3%, and F1-score of 91.9% compared to Naïve Bayes with an accuracy rate of 94.9%, precision of 91.7%, recall of 89.7%, and F1-score of 89.4%. This system is expected to support medical professionals by providing a faster and more accurate ARI diagnosis.

Actions (login required)

View Item