Determination of Student Graduation Using K-Nearest Neighbor (K-NN) Algorithm

Abstract

The education sector plays an important role in improving the overall development of a nation. To achieve this, schools must prioritize the quality of education by minimizing the number of students and implementing data mining processes. Data mining is used to collect large amounts of data, especially student data, which is continuously updated to ensure better information is available during the development process. The K-NN algorithm is a classification method that uses attributes and training samples to classify new data. However, this method has limitations, such as high computational costs and large dataset requirements. This research uses data from 60 students of SMK Swasta Anak Bangsa. The accuracy rate obtained using the K-NN algorithm to predict graduation based on knowledge, attitude, and attendance scores is 93.55%. The results show that K-NN can provide more accurate and effective data.