KLASIFIKASI MASA STUDI MAHASISWA MENGGUNAKAN ALGORITMA DECISION TREE (Studi Kasus : Mahasiswa Program Studi Informatika UMM)

In order to realize higher education efficiency, university managers are required to manage it professionally. One factor that influences efficiency is the student's length of study. Based on academic data from the Informatics study program at the University of Muhammadiyah Malang, many students have studied for more than four years. Length of study is one of the problematic factors for study program managers so that it can influence academic performance. In the academic environment, problems are still found based on data on graduating students who are unable to complete their studies within the specified time period. This can be used as a source of strategic information for study programs to classify the study period of graduate students using data mining techniques. Based on this background, this research needs to be carried out to apply data mining techniques to classify student study periods, namely by using one of the C4.5 Decision Tree algorithms. The method used is data analysis, data preprocessing, and classification process using the C4.5 Algorithm. In this research, 2 tests were carried out. The first is testing the C4.5 algorithm classification with the results obtained that the algorithm produces a classification accuracy of 82.05% with the root node or top node, namely the GPA value. For the second test, a comparison of the C4.5 algorithm was carried out with the previous algorithm, namely the ID3 algorithm, where the ID3 algorithm produced an accuracy of 81% with the same root node, namely the GPA value. This research shows that the C4.5 algorithm is the most effective and most appropriate algorithm to use in classifying data, with a fairly high prediction accuracy value compared to the previous algorithm, namely the ID3 Algorithm.