PREDIKSI POLA KELULUSAN MAHASISWA MENGGUNAKAN TEKNIK DATA MINING CLASSIFICATION EMERGING PATTERN

Abstract

Mercu Buana University is one of the private universities in Indonesia, especially in DKI Jakarta, which has a large number of students and a number of graduations. However, the University management has difficulty predicting a pattern and graduation rate from existing student data in each academic year. Most researchers use data mining techniques to find a regularity of patterns or relationships set on large data. In this paper, to predict patterns and analyze student graduation rates researchers use data mining by focusing on the classification process using emerging pattern algorithms on the timeliness of student studies. In this study the data used came from combined data between student master data and graduation data. The results of testing the data carried out by researchers in the data mining application produce graduation patterns with various variations according to the learning attributes used, namely gender, class, study program, lecture system and student GPA.

By using the results of testing this study, it is expected that the resulting data can help the management of the University as a basis for analysis in planning the teaching and learning process strategy to increase the graduation rate on time and as a support for the management of Mercu Buana University