Nurilawati, Meizapuspa Octakurnia (2026) SISTEM REKOMENDASI MINAT AKADEMIK MAHASISWA BERBASIS MACHINE LEARNING MENGGUNAKAN METODE K-MEANS DAN DECISION TREE. S1 thesis, Universitas PGRI Madiun.
![[img]](https://eprint.unipma.ac.id/style/images/fileicons/text.png) |
Text
Halaman Depan (Cover – daftar lampiran).pdf Download (626kB) |
|
Text
Abstrak (bahasa Inggris dan Indonesia).pdf Download (188kB) |
|
Text
BAB I_LAPORAN SKRIPSI MEIZAPUSPA OC_2205102003-21-30.pdf Download (203kB) |
|
Text
BAB II_LAPORAN SKRIPSI MEIZAPUSPA OC_2205102003-31-80.pdf Download (746kB) |
|
Text
BAB III_LAPORAN SKRIPSI MEIZAPUSPA OC_2205102003-81-143.pdf Restricted to Repository staff only Download (699kB) |
|
Text
BAB IV_LAPORAN SKRIPSI MEIZAPUSPA OC_2205102003-144-241.pdf Restricted to Repository staff only Download (3MB) |
|
Text
BAB V_LAPORAN SKRIPSI MEIZAPUSPA OC_2205102003-242-245.pdf Download (180kB) |
|
Text
Daftar pustaka_LAPORAN SKRIPSI MEIZAPUSPA OC_2205102003-246-255.pdf Download (207kB) |
|
Text
LAMPIRAN_LAPORAN SKRIPSI MEIZAPUSPA OC_2205102003-256-277.pdf Restricted to Repository staff only Download (2MB) |
Abstract
Determining students academic interests is an essential aspect of supporting competency development based on their individual potential. However, the process of identifying academic interests is often conducted subjectively, resulting in recommendations that do not fully represent students' characteristics and abilities. This study aims to develop a machine learning-based academic interest recommendation system by employing a hybrid approach that integrates the K-Means and Decision Tree CART algorithms to provide more objective recommendations. The study utilized questionnaire data collected from students of the Information Systems Study Program at Universitas PGRI Madiun, consisting of 13 indicators covering learning behavior, motivation, learning style, academic experience, professional experience, and course preferences. Data processing followed the Knowledge Discovery in Database (KDD) methodology, including data selection, data cleaning, data transformation, feature selection, pseudo-label generation using the Weak Supervision approach, clustering with the K-Means algorithm, and classification using the Decision Tree CART algorithm. The resulting model was then implemented in a web-based application using the Flask framework and Bootstrap. Model performance was evaluated using a Confusion Matrix, which showed an accuracy of 70.59%. The developed system successfully generated recommendations across six academic interest categories: Application Development, Data Science, UI/UX Design, Digital Business, Cyber Security & Networking, and Information Systems (Management & Analysis). These findings indicate that the proposed system can serve as a decision support tool or second opinion to assist students in selecting academic interests more objectively. Keywords: Recommendation System, Machine Learning, K-Means, Decision Tree CART, Weak Supervision, Academic Interest.
| Item Type: | Thesis/Skripsi/Tugas Akhir (S1) |
|---|---|
| Kata Kunci: | Kata kunci: Sistem rekomendasi, machine learning, K-Means, Decision Tree CART, Weak Supervision, minat akademik. Keywords: Recommendation System, Machine Learning, K-Means, Decision Tree CART, Weak Supervision, Academic Interest. |
| Subjects: | Q Science > QA Mathematics > QA75 Electronic computers. Computer science Q Science > QA Mathematics > QA76 Computer software |
| Divisions: | Fakultas Teknik > Sistem Informasi |
| Depositing User: | NURILAWATI OCTAKURNIA MEIZAPUSPA |
| Date Deposited: | 11 Aug 2026 07:48 |
| Last Modified: | 11 Aug 2026 07:48 |
| URI: | http://eprint.unipma.ac.id/id/eprint/8918 |
Actions (login required)
 |
View Item |
