Comparison of Predictive Machine Learning Models to Predict the Level of Adaptability of Students in Online Education

With the onset of COVID-19 pandemic, online education has become one most important options available to students around world. Although been widely accepted in recent years, sudden shift from face-to-face resulted several obstacles for students. This paper, aims predict level adaptability that have towards by using predictive machine learning (ML) models such as Random Forest (RF), K-Nearest-Neighbor (KNN), Support vector (SVM), Logistic Regression (LR) and XGBClassifier (XGB).The dataset used this paper was obtained Kaggle, which is composed a population 1205 high school college Various stages data analysis performed, including understanding cleaning, exploratory analysis, training, testing, validation. Multiple parameters, accuracy, specificity, sensitivity, F1 count precision, evaluate performance each model. The results shown all five can provide optimal terms prediction. For example, RF XGB presented best with an accuracy rate 92%, outperforming other models. In consequence, it suggested use these two prediction students' due their higher efficiency. Also, KNN, SVM LR models, achieved 85%, 76%, 67%, respectively. conclusion, show clear advantage achieving accuracy. These are line similar works ML techniques levels.