ARDIANTI, Mifta and Nurhayati, Oky Dwi and Warsito, Budi (2023) ANALISIS TEKNIK ENSEMBLE MACHINE LEARNING UNTUK PREDIKSI KINERJA AKADEMIK SISWA BERDASARKAN INTERAKSI SISWA PADA PEMBELAJARAN DARING. Masters thesis, UNIVERSITAS DIPONEGORO.
Text
Cover.pdf Download (369kB) |
|
Text
Bab I.pdf Download (85kB) |
|
Text
Bab II.pdf Download (697kB) |
|
Text
Bab III.pdf Restricted to Repository staff only Download (264kB) |
|
Text
Bab IV.pdf Restricted to Repository staff only Download (411kB) |
|
Text
Bab V.pdf Download (13kB) |
|
Text
Daftar Pustaka.pdf Download (158kB) |
Abstract
Institusi pendidikan saat ini menerapkan Adapive Learning Management System (LMS) sebagai sarana pembelajaran online. LMS dapat merekam sejumlah besar data perilaku siswa pada log LMS. Data perilaku ini dapat dikumpulkan dan digunakan untuk memprediksi kinerja belajar siswa. Sehingga, diperlukan analisis yang dapat mengubah sejumlah data yang tersimpan tersebut menjadi sebuah pengetahuan yang dapat meningkatkan kualitas pengajaran dan proses pembelajaran pada institusi pendidikan. Pada penelitian ini, mengusulkan model prediksi kinerja belajar siswa menggunakan ensemble machine learning berdasarkan variabel yang berhubungan dengan interaksi siswa pada LMS. Pemodelan dilakukan dengan menerapkan algoritma ensemble machine learning dengan tiga jenis model yaitu ; bagging, boosting dan voting yang menggabungkan algoritma logistic regression dan support vector machine. Hasil penelitian ini adalah ensemble machine learning dapat diterapkan untuk memprediksi kinerja belajar siswa. Nilai akurasi tertinggi dari model bagging dengan algoritma random forest yaitu sebesar 80,21%, precision 0,7952, recall 0,7988 dan f-measure sebesar 0,7955. Ensemble machine learning dapat diterpkan untuk prediksi kinerja siswa berdasarkan perilaku belajar.
Kata kunci : ensemble machine learning, LMS, educational data mining, prediksi kinerja siswa
Educational institutions are currently implementing Adaptive Learning Management System (ALMS) as a platform for online learning. The ALMS can record a large amount of student behavior data on the LMS log. This behavioral data can be collected and used to predict students' learning performance. Thus, an analysis is needed that can transform the amount of stored data into knowledge that can improve the quality of teaching and learning process in educational institutions. In this research, a student learning performance prediction model using ensemble machine learning is proposed based on variables related to student interactions on the ALMS. Modeling is done by applying the ensemble machine learning algorithm with three types of models namely; bagging, boosting and voting which combines logistic regression and support vector machine algorithms. The result of this research is that ensemble machine learning can be applied to predict student learning performance. The highest accuracy of bagging model with random forest algorithm is 80.21%, precision 0.7952, recall 0.7988 and f-measure of 0.7955.
Keyword : Ensemble Machine Learning, LMS, Educational Data Mining, Student Performance Prediction
Item Type: | Thesis (Masters) |
---|---|
Uncontrolled Keywords: | ensemble machine learning, LMS, educational data mining, prediksi kinerja siswa |
Subjects: | Sciences and Mathemathic |
Divisions: | Postgraduate Program > Master Program in Information System |
Depositing User: | ekana listianawati |
Date Deposited: | 19 Sep 2023 03:34 |
Last Modified: | 19 Sep 2023 03:34 |
URI: | https://eprints2.undip.ac.id/id/eprint/16385 |
Actions (login required)
View Item |