Analisis Klasifikasi Konsentrasi Mahasiswa Menggunakan Algoritma K-Nearest Neighbor

Abstract

Hamzanwadi University located in West Nusa Tenggara boasts a Faculty of Engineering, which plays a pivotal role in delivering top-notch higher education. This faculty offers four highly coveted programs—Informatics Engineering, Information Systems, Computer Engineering, and Environmental Engineering—that attract a significant number of students. The increase in enrollment in these programs underscores the faculty's success, particularly owing to the promising job opportunities in these fields. Nevertheless, students often encounter intricate challenges when selecting their area of specialization that resonates with their interests and capabilities. In addressing this concern, the Faculty of Engineering at Hamzanwadi University provides diverse concentration options like Data Science, RPL, and Multimedia. To aid students in making informed decisions regarding their study concentration, this study employs the K-Nearest Neighbor (KNN) algorithm to analyze the classification of student concentrations. This research adopts an experimental approach and utilizes data collection methods such as observation, interviews, and surveys. The dataset comprises seven attributes including NIM, Gender, GPA, Data Science Course Grade, RPL Course Grade, and Multimedia Course Grade, processed using the KNN algorithm through Google Colab. The research outcomes reveal that with k=2 and 8-fold cross-validation, the achieved accuracy stands at 67%.