Career planning matters: Intelligence-based career path predictions using data mining models - A longitudinal study

Abstract

It is essential for students to plan their careers because selecting the right career path shapes a person’s life. In such disciplines as computer science, information technology, and software engineering, assisting students toward appropriate employment is even more helpful. Thus, throughout the student’s education, they must evaluate their strengths to determine which professional sector corresponds to their abilities. It is for this reason that this research introduces an intelligencebased career recommendation system that will incorporate the analysis of such factors as the student’s academic performance, economic status, and demographic features to use data mining models in determining the best prospective career path to offer the student a more transparent and more informed vision and course to set on in the future. Three key aspects were addressed: first, one or another model and classifier for assessing the impact of pre-university education on the choice of a profession and, consequently, the selection of technologies were used. Second, these models accurately forecast students' careers using core courses, CGPA, and FYP data. Third, socioeconomic or demographic data was incorporated into the prediction to make it more accurate. Regarding the method of class distribution balancing, the Synthetic Minority Oversampling Technique (SMOTE) approach was used. The study reveals that variables specifying preuniversity education directly impact students’ career choices and that, employing data mining techniques, career choices could be forecasted considering academic performance and other related factors.