Employment Recommendation System for Graduates Using Machine Learning

Abstract

Employee selection is one of the human resources (HR's) challenging tasks that require a decision support system to help them process the task quickly. Generally, establishing a solid and stringent selection process assists the department and organization in reaching its goals, saving time, and redirecting effort to more important things. In this study, a recommendation system that can accurately identify and classify the best candidates for specific positions is presented. This system is based on using supervised machine learning techniques to match job seekers with suitable job opportunities. In particular, this paper investigates various supervised machine learning algorithms; Decision Tree, Support Vector Machine, and Random Forest to predict and suggest jobs to graduates based on their educational achievements and work history. Using a set of graduates’ data, the system's results were evaluated. The results show that the fine-tuned random forest does a better prediction than the other algorithms at making accurate and personalized job recommendations, with an accuracy of 98.43%. The importance of features was also investigated, and it was found that the Secondary School Percentage, the Higher Secondary School Percentage Degree Percentage, and the Student Percentage in MBA were the most important features. This means that this model depends heavily on the students' degrees.