Stress Level Classification Towards Student Learning Outcomes Using Support Vector Machine
Stress Level Classification Towards Student Learning Outcomes Using Support Vector Machine

Summary/Abstract: This research aims to group stress factors that influence student learning outcomes in higher education and classify the level of student learning outcomes. This study conducted the factor analysis to group the psychological stress variables of students. The factor analysis results were then used as inputs to the Support Vector Machine (SVM) classifier while the student learning outcomes served as outputs. Then proceed with the training and testing of SVM to be able to classify student learning outcomes correctly. The study results indicated that 36 stress variables could be grouped into 7 stress factors using factor analysis, namely conflict, learning readiness, participation, workload, fairness, focusing, and suitability. Four levels were used to categorize the learning results of students, such as very good, good, fair, and poor. The accuracy and precision of the SVM classification values were 0.89 and 0.70, respectively. The SVM classification results showed better results in comparison to comparable methods of classification like Decision Trees, Random Forests, K-Nearest Neighbor (K-NN), Linear Discriminant Analysis (LDA), and Gaussian Naive Bayes (GNB). The implications of this research can be used to classify student learning outcomes based on stress factors that accompany them during their studies.

• Details
• Contents