Mining Student Behavioral Concern Through Referrals Using K-Means Clustering

Abstract

Monitoring students' behavior is one the main concerns faced by higher education institutions nowadays. Several procedures were taking into consideration to do the former statement and one of it is through referrals. The research aimed to cluster student behavior based on the referrals using K-Means algorithm. In this paper, students were clustered into two groups according to gender and year level. The result showed that the primary reasons of referrals among the male students were absences, tardiness, poor academic performance and on probation. On the other hand, for the female was misconduct. In terms of the year level of the students, it showed that the primary reasons of referrals among the first and second year students were absences, tardiness, poor academic performance and on probation. While for the third and fourth year students was misconduct.

Metode k-means clustering untuk mengukur tingkat kedisiplinan pegawai (studi kasus di pemerintah kabupaten padang pariaman)

“…Melakukan penentuan mata kuliah peminatan mahasiswa [15]. Mengelompokan prilaku mahasiswa berdasarkan ketidakhadiran, keterlambatan, nilai akademik [16]. Penelitian lainnya yang menggunakan algoritma k-means clustering dalam menentukan kepadatan permintaan dari berbagai jenis pekerjaan [17], melakukan zonasi suhu air waduk [18].…”