Analisis Performa Algoritma K-Nearest Neighbor dalam Klasifikasi Penyakit Tumor Otak

Abstract

Brain tumor disease poses a significant health challenge globally, including in Indonesia. Detecting brain tumors early is crucial for effective treatment. In this study, we investigated the performance of the K-Nearest Neighbor (KNN) algorithm in classifying brain tumor disease using brain image data. Our findings reveal that the choice of K value significantly impacts the KNN algorithm's performance. The highest accuracy of 81% was achieved with K=3, while the lowest accuracy of 66% occurred at K=7. On average, across all scenarios, the accuracy was 72.8%. These results underscore the importance of selecting the appropriate K value for optimal classification accuracy in brain tumor disease using the KNN algorithm.