Kualitas lulusan dari sebuah Perguruan Tinggi salah satunya dapat dilihat dari lama studi mahasiswa. Selain itu lama studi mahasiswa menggambarkan tingkat capaian mahasiswa dalam pendidikannya. Lama studi juga sangat berpengaruh pada kualitas program studi karena lama studi mahasiswa merupakan salah satu kriteria penilaian akreditasi. Seringkali masalah yang dihadapi oleh suatu Perguruan Tinggi adalah banyaknya mahasiswa yang menyelesaikan pendidikannya lebih dari jangka waktu yang ditetapkan. STMIK Bina Nusantara Jaya Lubuklinggau juga mengalami hal tersebut. Untuk mengantisipasi hal tersebut perlu adanya prediksi lama studi mahasiswa karena lama studi mahasiswa menjadi salah satu hal yang penting yang perlu diperhatikan bagian program studi dalam suatu Perguruan Tinggi. Penelitian ini berkontribusi secara teoretis dalam implementasi data mining untuk memprediksi lama studi mahasiswa.Penelitian ini menerapkan preprocessing data untuk memperoleh data dengan kualitas baik sebelum dilakukan proses mining menggunakan metode K-Nearest Neighbor dan Decision Tree pada Tools RapidMiner, kedua metode divalidasi menggunakan K-Fold Cross Validation (dengan 10 kali iterasi/pengulangan) dan Confusion Matrix digunakan untuk memvalidasi nilai akurasi hasil prediksi. Nilai akurasi yang paling tinggi dari hasil penerapan kedua metode akan direkomendasikan untuk menyelesaikan masalah prediksi lama studi mahasiswa. Dari hasil penelitian diperoleh nilai akurasi metode Decision Tree (60,38%) lebih baik jika dibandingkan dengan nilai akurasi metode K-Nearest Neighbor (53,08%).
Sejarah Artikel:The inability of students to complete their studies on time is faced by most of higher education institution. STMIK Bina Nusantara Jaya Lubuklinggau is one of those which is experienced with this matter. In most cases, the students could complete their studies longer than the expected duration. From 162 students of Sistem Informasi study program in the year 2013 and 2014, there were 117 students completed their studies on time, while 45 students were late. As a result, it could prevent new students from joining the institution since the limited student capacity. This study deploys data mining technique in predicting the graduation status of students on time. First, preprocessing is used to obtain a good dataset. Secondly, the data is processed to obtain a set of prediction. In this step, two mining algorithm were applied -Naive Bayes classifier and C4.5 algorithm to be knowing the performance of the two methods, the method has a greater accuracy value will be recommended to solving the problem of prediction of students graduation at STMIK Bina Nusantara Jaya Lubuklinggau. Thirdly, the result then was validated using K-Fold Cross Validation technique. Finally, the Confusion Matrix is deployed to ensure the accuracy of the prediction. The results indicate that the C4.5 Algorithm method can be used to predict student graduation status with an accuracy rate of 79,08% while the accuracy rate of the Naive Bayes Classifier method is only 78,46%. The dominant factor is IPK-S4 variable.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.