DETEKSI DINI KANKER PARU PADA PASIEN MENGGUNAKAN METODE K-MEANS CLUSTERING

Abstract

Abstract
Lung cancer is one of the leading causes of death worldwide, with high mortality rates due to delays in diagnosis.
Lung cancer symptoms are often only detected at an advanced stage, making early detection crucial to improving
the chances of recovery. This study aims to develop an early detection method for lung cancer in patients using
the K-Means Clustering algorithm based on the symptoms experienced. The data used consists of 309 patient
samples with 16 attributes reflecting health conditions such as age, smoking habits, coughing, and shortness of
breath. The K-Means method was applied to group patient data into several clusters based on symptom similarities,
helping to identify symptom patterns associated with the risk of lung cancer. The results show that this algorithm
successfully divided patients into two main clusters indicating different risk levels. The use of the K-Means
Clustering method has proven effective in aiding early diagnosis of lung cancer and has the potential to
significantly improve detection accuracy and accelerate the medical treatment process.
Keywords: Detection, Lung Cancer, K-Means, Cluster, Symptom