Siregar, Fitri Yanti and Sikumbang, Hengki and Asri, Yessy (2025) PENERAPAN ALGORITMA RANDOM FOREST UNTUK KLASIFIKASI TINGKAT KEPUASAN MAHASISWA BERDASARKAN ASPEK UI/UX PADA PLATFORM LEARNING MANAGEMENT SYSTEM (LMS): STUDI KASUS DI ITPLN. Diploma thesis, ITPLN.
202131091_Fitri Yanti Siregar_Revisi_Skripsi_Fitriyanti Siregar.pdf
Restricted to Registered users only
Download (4MB)
Abstract
Perkembangan teknologi informasi mendorong perguruan tinggi untuk menggunakan Learning Management System (LMS) sebagai media pendukung pembelajaran daring yang fleksibel dan terstruktur. LMS memfasilitasi pengunduhan materi, peninjauan nilai, akses jadwal kuliah, serta interaksi antara dosen dan mahasiswa melalui satu platform terpadu. Penelitian ini bertujuan untuk mengklasifikasikan tingkat kepuasan mahasiswa terhadap LMS di Institut Teknologi PLN (ITPLN) menggunakan algoritma Random Forest, yang dikenal mampu menghasilkan prediksi akurat dan mengelola data dengan banyak variabel. Data penelitian diperoleh dari 300 mahasiswa aktif melalui kuesioner berbasis skala Likert lima tingkat. Setelah dilakukan pra-pemrosesan, label target dibagi menjadi dua kategori, yaitu “Puas” dan “Kurang Puas”. Model awal dengan 45 fitur menghasilkan akurasi sebesar 80,00%, sedangkan seleksi 10 fitur utama meningkatkan akurasi menjadi 85,00%. Hyperparameter tuning melalui GridSearchCV menghasilkan model terbaik dengan akurasi 98,33% dan F1-score 98,80%. Evaluasi tambahan menggunakan 5-Fold Cross Validation memberikan rata-rata akurasi 82,33%, yang membuktikan bahwa kombinasi seleksi fitur dan penyetelan parameter mampu meningkatkan akurasi sekaligus kemampuan generalisasi model secara signifikan.
The development of information technology has encouraged higher education institutions to adopt Learning Management Systems (LMS) as flexible and structured platforms for online learning. LMS facilitates downloading course materials, reviewing grades, accessing class schedules, and enabling interactions between lecturers and students through a single integrated platform. This study aims to classify student satisfaction levels with the LMS at Institut Teknologi PLN (ITPLN) using the Random Forest algorithm, which is known for its high predictive accuracy and ability to handle datasets with many variables. Data were collected from 300 active students through a five-point Likert scale questionnaire. After preprocessing, the target labels were categorized into “Satisfied” and “Less Satisfied.” The initial model using 45 features achieved an accuracy of 80.00%, while selecting the top 10 features increased the accuracy to 85.00%. Hyperparameter tuning using GridSearchCV yielded the best model with an accuracy of 98.33% and an F1-score of 98.80%. Additional evaluation using 5-Fold Cross Validation resulted in an average accuracy of 82.33%, confirming that the combination of feature selection and parameter tuning significantly enhances both the model’s accuracy and its generalization capability.
| Item Type: | Thesis (Diploma) |
|---|---|
| Uncontrolled Keywords: | Random Forest, Klasifikasi, LMS, Machine learning, Kepuasan Mahasiswa Random Forest, Classification, LMS, Machine learning, Student Satisfaction |
| Subjects: | Skripsi Bidang Keilmuan > Teknik Informatika |
| Divisions: | Fakultas Telematika Energi > S1 Teknik Informatika |
| Depositing User: | Sudarman |
| Date Deposited: | 13 Oct 2025 01:59 |
| Last Modified: | 13 Oct 2025 01:59 |
| URI: | https://repository.itpln.ac.id/id/eprint/2075 |
