Towards enthusiasm prediction of Portuguese school's students towards higher education in realtime

An experimental study has been conducted to predict the future enthusiasm of language course students to take admission into higher education. Binary classification is resolved with the application of machine learning techniques on the standard dataset of two Portuguese schools with language subject available on the UCI website. Three supervised machine learning algorithms likewise Support vector machine (SVM), Multilayer Perceptron (ANN), and Random Forest (RF) are modeled on 649 instances and 33 features. The outcomes of the study may predict the student's willingness to go for their further higher education after finishing their schooling. Their willingness was based on features such as student grades, demographic, personal, social and school. The results of each experiment proved that the RF predictive model outperformed others in the prediction of student enthusiasm with most accuracy of 96.31 % using cross-validation. Also, the Mathews correlation proved that all features in each dataset are highly positively correlated with enthusiasm. The statistical t-test also proved a significant difference between the accuracy of the SVM and the ANN. The cross-validation method also proved robust and reliable as compared to the training ratios. The RF enthusiasm predictive model is highly acclaimed to be developed as a web application in school to recognize the student enthusiasm towards higher education.