Penerapan Data Mining Menggunakan Algoritma C4.5 untuk Klasifikasi Penyakit Paru-Paru

Abstract

Lung disease remains one of the leading causes of morbidity and mortality in Indonesia. Factors such as smoking habits, unhealthy lifestyles, and low awareness of respiratory health have significantly contributed to the increasing prevalence of this condition. This study aims to classify lung health conditions using the Decision Tree C4.5 algorithm, a data mining technique widely applied in medical analysis. The dataset consists of 132 respondents with various attributes, including age, gender, smoking habits, sleep patterns, and medical history. The model was validated using the percentage split method, A total of 80% of the data was allocated for training the model, while the remaining 20% was used for testing its performance.The results show that the C4.5 algorithm successfully classified lung disease risk into two categories, “Yes” (at risk) and “No” (not at risk), with a high level of accuracy. The model effectively identified the key factors contributing to lung disease risk, such as smoking habits, late-night activity, age, and insurance ownership. These findings confirm that the Decision Tree C4.5 algorithm is a reliable and efficient tool for the early detection of respiratory diseases and can support data-driven decision-making in the healthcare field.