Real-time classification of national and international students for ICT and mobile technology: an experimental study on Indian and Hungarian University

Abstract

Abstract This paper focused on the classification of student national-level status such as national and international. A primary survey was conducted in the academic year 2017-2018 to analysis the circumstances of trending ICT and Mobile Technology (MT) in Indian and Hungarian higher education. The main objective was to identify the student’s answers provided in the survey based on their national-level status. For the classification tasks, we used Logistic regression (LR), Support Vector Machine, Multilayer perceptron (MLP) and Random Forest (RF)on both balanced and unbalanced datasets with K-fold Cross-Validation (KCV), Leave One Out (LOO), and Hold Out (HO) methods. Also, Xtreme Gradient Boosting (XGB) classifier was also implemented to enhance the classification accuracy of existing classifiers. The findings of the study showed that the XGB classifier outperformed others with the highest accuracy of 95% with 18 significant features. Also, class balancing improved significantly the accuracy of classification. Further, the authors recommended this predictive model to be implemented as a real-time function utility on the website of the university.