Klasifikasi Tingkat Kelulusan Mahasiswa Menggunakan Algoritma K-Nearest Neighbor (K-NN) Pada Data Akademik Perguruan Tinggi

Efendi, Davina Rizky
Irmayani, Deci
Sihombing, Volvo

Journal of Computer Science and Information Systems (JCoInS)
Vol 6, No 3: JCoInS | 2025

Higher education is an important factor in scoring quality human resources, where one indicator of success is the graduation rate of students on time. This study aims to classify the graduation rate of students using the algorithm K-Nearest Neighbor (K-NN) based on academic data which includes GPA, number of credits, frequency of repetition of courses, and attendance. The results of the classification showed that 30% of students successfully graduated on time, while the rest had delays. With the k-NN approach, it is expected that this model can help universities in predicting student graduation more accurately and optimizing academic interventions to improve graduation efficiency.