A Review of Decision Tree Algorithms and Their Applications in Academic Guidance: A Systematic Literature Review

This systematic literature conducted according to the PRISMA framework, provides a comprehensive analysis of research conducted between ۲۰۲۰ and ۲۰۲۵ on the application of decision tree algorithms in academic guidance. Following a rigorous search that identified ۵۰۰ articles, a multi-stage screening process led to the inclusion of ۷۸ high quality studies. This review synthesizes the theoretical foundations, advancements, applications, challenges, and future directions of these algorithms. Findings indicate that decision trees, due to their high interpretability, transparency, and hierarchical structure, serve as effective tools in educational decision-making. They are particularly prominent in student performance prediction and course recommendation systems, where they have demonstrated high accuracy (۸۲-۹۵%) and contributed to a ۱۵% or greater increase in course completion rates. Despite their technical success, emerging challenges include ethical concerns such as algorithmic bias, data privacy, and scalability. Current research trends are shifting toward hybrid models, explainable AI (XAI), and real-time, scalable systems. The review concludes that decision trees have the potential to revolutionize personalized education. To fully realize this potential, future research must simultaneously focus on three pillars: (۱) enhancing technical performance through interpretable and hybrid models, (۲) ensuring ethical principles such as fairness, transparency, and privacy, and (۳) improving practical applicability via real-time and scalable systems. Only through this comprehensive approach can academic guidance systems evolve into intelligent, equitable, and trustworthy tools that genuinely support the success of all students.

Keywords:

Decision Tree , Machine Learning , Academic Guidance , Student Performance Prediction , Course Recommendation , Interpretability