Application of Classification Data Mining Technique for Pattern Analysis of Student Graduation Data with Emerging Pattern Method

Abstract

Data mining has been applied in various fields of life because it is very helpful in extracting information from large data sets. Student graduation data is one example of data that can be extracted for information and become a recommendation. This study used a classification data mining technique to extract information from the student graduation data. The classification technique used was the Emerging Pattern method to search for patterns in the student graduation data. The data in this study were graduation data for students of the Statistics Study Program, Faculty of Mathematics and Natural Sciences, Tanjungpura University, from 2013-2018. The sample data used amounted to 186 records. Attributes used in this study include as many as four attributes, including gender, batch, GPA, and TUTEP scores. This research began by finding the class and frequency values obtained. It was continued by calculating each item set's support, growth rate, and confidence values. This study obtained the highest confidence value among all the attributes owned, namely 91% in the 2013 batch itemized list and the 2018 batch. Female students dominated the class attribute. TUTEP dominated the TUTEP value attribute with a score of 425, and the GPA attribute of 3.51-4.00 dominated the class with a confidence value of 60%.