An Ensemble Model for Predicting Career Paths of Sri Lankan IT Undergraduates

Abstract

Career trajectories of Information Technology (IT) undergraduates are critical for tailoring academic programs to align with industry demands and support students' professional aspirations. Rapid changes in the IT industry present both opportunities and challenges for new graduates entering the job market. Despite the availability of information and education, many IT undergraduates struggle to secure jobs that align with their career aspirations and abilities. Traditional career development frameworks often fail to provide accurate and tailored predictions due to their inability to address industry-specific requirements, technical skills, and soft skill demands. This mismatch contributes to graduate unemployment, skills imbalances, and career insecurity. To address these challenges, this research proposes a Machine Learning (ML)-based model for predicting suitable career paths for Sri Lankan IT undergraduate students by analyzing their skills, educational background, and internship experiences. The study focuses on six distinct career fields: Software Engineering, UI/UX Design, Quality Assurance, Business Analysis/Project Management, Data Science/Artificial Intelligence, and Networking/DevOps/System Administration. The authors identified 31 attributes through an extensive literature survey and expert opinions. To implement the ML models, a dataset of 820 employees working in the IT industry was obtained and preprocessed. The dataset was then prepared for the application of ML models. This research employs eight different ML algorithms to predict the career paths of IT undergraduates and compares their effectiveness. The Ensemble model demonstrates superior performance compared to other models, with accuracy, precision, recall, and F1-score of 91.39%, 92.47%, 91.39%, and 91.48%, respectively.