Please use this identifier to cite or link to this item: http://repository.ipb.ac.id/handle/123456789/170683
Title: Pengembangan Model Pohon Keputusan C5.0 dengan Teknik Grafting untuk Identifikasi Keberlanjutan Penerima KIP Kuliah
Other Titles: Development of a C5.0 Decision Tree Model with the Grafting Technique for Identifying the Sustainability of KIP Kuliah Recipients
Authors: Wijaya, Sony Hartono
Ardiansyah, Firman
Rohman, Nur
Issue Date: 2025
Publisher: IPB University
Abstract: Program Kartu Indonesia Pintar Kuliah (KIPK) memberikan bantuan pendidikan bagi mahasiswa dengan keterbatasan ekonomi. Namun, evaluasi penerima di Institut Pertanian Bogor (IPB) selama ini masih terbatas pada pengumpulan data daftar ulang tanpa analisis mendalam, sehingga berpotensi menimbulkan ketidaktepatan sasaran. Penelitian ini bertujuan mengembangkan model klasifikasi keberlanjutan penerima KIPK dengan algoritma C5.0 yang dimodifikasi menggunakan teknik grafting. Data penelitian diperoleh dari data daftar ulang penerima KIPK di IPB. Model dasar dibangun menggunakan C5.0, kemudian dimodifikasi dengan teknik grafting untuk mengakomodasi penambahan variabel baru tanpa membangun ulang model. Hasil penelitian menunjukkan bahwa teknik grafting meningkatkan fleksibilitas model dalam menghadapi perubahan kebijakan. Penelitian ini juga menghasilkan aplikasi berbasis R Shiny untuk memvisualisasikan model, memilih variabel, dan menambahkan aturan baru sesuai kebutuhan pengelola beasiswa. Temuan ini menunjukkan bahwa pengembangan model C5.0 dengan teknik grafting efektif dalam mendukung identifikasi keberlanjutan penerima KIPK secara adaptif.
The Kartu Indonesia Pintar Kuliah (KIPK) program provides educational assistance for students with economic limitations. However, the evaluation of recipients at Institut Pertanian Bogor (IPB) has been limited to collecting re-registration data without in-depth analysis, potentially leading to inaccurate targeting. This research aims to develop a classification model for the sustainability of KIPK recipients with a modified C5.0 algorithm using a grafting technique. The research data was obtained from the re-registration data of KIPK recipients at IPB. The basic model was built using C5.0, then modified with grafting techniques to accommodate the addition of new variables without rebuilding the model. The results showed that the grafting technique increased the flexibility of the model in the face of policy changes. This research also produced a Shiny R-based application to visualize the model, select variables, and add new rules according to the needs of scholarship managers. These findings show that the development of C5.0 models with grafting techniques is effective in supporting the identification of the sustainability of KIPK recipients adaptively.
URI: http://repository.ipb.ac.id/handle/123456789/170683
Appears in Collections:UT - Computer Science

Files in This Item:
File Description SizeFormat 
cover_G6401201053_8756eddfa6854af3bb387bba4b1cbc05.pdfCover346.33 kBAdobe PDFView/Open
fulltext_G6401201053_f82464e3e8624f7faa52d2f8d4e0c5a3.pdf
  Restricted Access
Fulltext1.27 MBAdobe PDFView/Open
lampiran_G6401201053_f12e3471ddf24f5285105c9e620d9ea7.pdf
  Restricted Access
Lampiran347.39 kBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.