WINDY JUNITA SARI, - (2025) PEMILIHAN ATRIBUT TERBAIK DENGAN REDUKSI DIMENSI DAN SELEKSI FITUR BERBASIS MACHINE LEARNING UNTUK KLASIFIKASI KINERJA SISWA. Skripsi thesis, Universitas Islam Negeri Sultan Syarif Kasim Riau.

Preview	Text WINDY JUNITA SARI.pdf Download (13MB) | Preview
	Text (BAB IV) BAB 4.pdf - Published Version Restricted to Repository staff only Download (770kB)

Abstract

Education plays an important role in shaping the intellectual and character of the nation’s next generation. However, poor academic performance of students is a big challenge, especially related to student retention and dropout risk. This study aims to evaluate the performance of Machine Learning Algorithms namely K-NN, LGBM, and XGB, and analyze the effect of dimensionality reduction using PCA and feature selection with RFE on student performance prediction accuracy. The research dataset consists of 395 student samples with demographic, social, and academic attributes. The results show that XGB has the best performance with 98,32% accuracy and is able to predict all classes with perfect 100% accuracy. LGBM and K-NN achieved 94,87% and 93,88% accuracy respectively. The best attributes that affected student performance were found in the “Highly Prioritized” category, which included Studytime, Famsup, Famrel, and Health. Although PCA slightly degraded the model performance, feature selection with RFE significantly improved accuracy. This study concludes that proper algorithm selection and focus on relevant attributes can improve prediction accuracy and efficiency, making an important contribution to the development of more effective educational prediction systems.

Actions (login required)

View Item