Visa Prediction for Higher Studies Using Machine Learning

Abstract

Computer science is arguably one of the most common fields across both Bangladesh and the world today. It is obvious that a statistically significant percentage of learners struggle to achieve the peak of this discipline due to the lack of skill in this discipline. Without a doubt, one of the most popular studies is going abroad for higher studies. It is really necessary for students to choose the correct path before applying for a higher education visa in order to succeed. In this work, we predict the visa for higher studies based on student’s information. Then we process those data (like; cleaning, transformation, integration, standardization, feature selection). Later we used different classification techniques i.e. C4.5 (j48), K-NN, Naive Bayes, Random Forest, SVM, Neural Network to classify these profiles. Based on the result analysis, it has been found that accuracy and other factors of a confusion matrix for Random Forest classifiers are more cogent than others. We also find out the attributes upon which a student’s visa accepted depends mostly. Therefore, the GRE score, Undergraduate CGPA, are two of the most important factors to determine success in the visa approval for higher studies.