Open AccessImplementasi Algoritma Clustering K-Means Untuk Mengelompokkan Mahasiswa Baru Yang Berpotensi (Studi Kasus: STMIK Budi Darma)
Author(s) -
Efori Buulolo,
Rian Syahputra
Publication year - 2019
Publication title -
prosiding seminar nasional riset information science (senaris)
Language(s) - English
Resource type - Journals
ISSN - 2686-0260
DOI - 10.30645/senaris.v1i0.3
Subject(s) - cluster analysis , ranking (information retrieval) , class (philosophy) , selection (genetic algorithm) , computer science , mathematics education , psychology , machine learning , artificial intelligence
Every year STMIK Budi Darma accepts new students through the selection test path. The process of student admission at STMIK Budi Darma begins with socialization to the community or school, registration, selection examinations, announcement of exam results, Re-registration, Campus Recognition System (CRS) and the lecture process. The value of the selection test results, the average National exam score, the mathematics National exam scores and the English National exam scores are combined and used in ranking prospective students who are accepted based on the capacity of the study program. So far, STMIK Budi Darma lecturers sometimes find it difficult to carry out classroom actions because each student has unequal potential so the learning outcomes are not optimal. One of them is caused by grouping new students into one class based on ranking, not based on the potential of each new student. In order to group prospective new students based on their potential, K-Means clustering algorithm is used. The basis of grouping with the K-Means clustering algorithm is based on the closest distance values of numerically based criteria with output in the form of clusters. The cluster that has been formed is used as the basis for grouping new students into one class.