Penerapan Metode CRISP-DM untuk Prediksi Kelulusan Studi Mahasiswa Menempuh Mata Kuliah (Studi Kasus Universitas XYZ)

Abstract

XYZ University as the leading private universities in Indonesia is required to improve its quality of education to produce quality graduates. To produce quality graduates, students’ success in their study is very influential. Student study success can be judged from several aspects of assessments in the lecture. For information systems courses, student success can be viewed in several student assessment aspects such as assignments, Mid-Semester Exam, and End-Semester Exam. More in-depth analysis on the value of the portion of the pattern of student needs in order to know the hidden patterns and to discover the knowledge of students will predict course passing success. This pattern is expected to help increase the success of students study. This research used CRISP-DM methods as standard processes for data mining that can be applied to the general problem-solving strategies on business or to other research units. Election algorithm as an algorithm C4.5 decision tree is used to facilitate the establishment. The results of this research study is the prediction of the pattern of graduation students in taking the course. This pattern is generated from the decision tree is expected to be a reference for graduate studies students take a course in information systems courses XYZ University. Keywords - CRISP - DM, C4.5 algorithm, graduation studies, student, and course