Agustinus, Maks (2018) Perancangan Sistem Peringatan Dini Drop Out di MMT ITS Menggunakan Metode Klasifikasi Naive Bayes. Masters thesis, Institut Teknologi Sepuluh Nopember.
Preview |
Text
9116205335-Master_Thesis.pdf - Accepted Version Download (4MB) | Preview |
Abstract
Kejadian dropout bukan hanya merugikan bagi mahasiswa yang mengalaminya, tetapi terutama bagi negara dan institusinya. Karena singkatnya masa studi normal, kejadian dropout seolah tiba-tiba. Sehingga dibutuhkan sistem peringatan dini yang mampu memberikan peringatan sedini mungkin guna keberhasilan tindakan pencegahan. Metode-metode klasifikasi dalam data mining, menjadi alternatif solusi dalam proses klasifikasi drop out pada sistem peringatan dini ini.
Dari beberapa metode klasifikasi yang baik menurut beberapa penelitian terbaru, Naive Bayes dipilih sebagai alternatif. Metode klasifikasi Naive Bayes digunakan untuk menghasilkan model dalam sebuah sistem peringatan dini dropout untuk kemudian digunakan dalam penentuan klasifikasi dropout seorang mahasiswa. Output sistem peringatan dini ini adalah daftar mahasiswa yang terklasifikasi dropout. Daftar ini dapat digunakan oleh akademik untuk tindakan pencegahan.
Proses krusial pembuatan sistem peringatan dini ini ada pada pemilihan variabel prediktor yang cocok digunakan dalam metode klasifikasi Naive Bayes. Rancangan yang sesuai dibutuhkan untuk mewujudkan implementasi metode klasifikasi Naive Bayes dalam sistem peringatan dini drop out.
======================================================================================
Impact of dropout event is not only affect the student, but also affect the country and the institution. Because of short of study time at magister program, dropout event looks appear suddently. In this situation, early warning system is needed here so that it gives an alert as early as possible for prevention action. Data mining classification methods become alternative solution in dropout classification process.
From some good classification methods from some newest research, Naive Bayes is chosen as an alternative method that will be implemented. Naive Bayes is used to generate a classification model for drop out early warning system. The output of this system will be a list of student that are classified as dropout. This list may be used by academic unit for prevention.
Crucial process in building a dropout early warning system is in choosing the suitable predictor variables for Naive Bayes method. Proper design is needed to implement Naive Bayes classification method in a dropout early warning system.
| Item Type: | Thesis (Masters) |
|---|---|
| Additional Information: | RTMT 005.276 Agu p-1 3100018074200 |
| Uncontrolled Keywords: | Early Warning System, Dropout, Data Mining, Naive Bayes |
| Subjects: | Q Science > QA Mathematics > QA76.9.D343 Data mining. Querying (Computer science) T Technology > T Technology (General) > T58.62 Decision support systems |
| Divisions: | Faculty of Business and Management Technology > Management Technology > 61101-(S2) Master Thesis |
| Depositing User: | Maks Agustinus |
| Date Deposited: | 05 Mar 2018 04:38 |
| Last Modified: | 03 Jul 2020 03:29 |
| URI: | http://repository.its.ac.id/id/eprint/49712 |
Actions (login required)
 |
View Item |