PENERAPAN DATA MINING PADA PENERIMAAN MAHASISWA BARU DENGAN ALGORITMA K-MEANS CLUSTERING

Abstract

This paper aims to grouping data using Clustering method with k-means algorithm to find potential majors and type of schools that produce feature students who have a good GPA score in semester 1 and semester 2 at Politeknik STMI Jakarta. Dataset from academic data for 2017-2020 has been processed with Rapid Miner showing that in Automotive Business Administration study program there are 3 clusters of students where cluster 0 marked as best cluster is dominated by high school students majoring in Science and Social Sciences. Automotive Industry Information System study program produces 2 clusters of students where cluster 0 marked as best cluster is dominated by high school students majoring in science and vocational high school majoring in mechanical engineering. Automotive Industrial Engineering study program produces 2 clusters of students where cluster 1 marked as best cluster is dominated by high school students majoring in science. Polymer Chemical Engineering study program produces 6 student clusters where cluster 4 marked as best cluster which all come from high school students majoring in science.