A Decision Tree-Based Advisory Recommendation System for Dental Students

Abstract

This study aims to enhance dental education by developing a model that matches dental students with suitable advisors using Data Mining Classification Techniques. Inadequate guidance from mismatched advisors can hinder students' academic and practical performance, leading to suboptimal educational outcomes. Questionnaires were used to analyze the relationship between students' expectations and advisor behaviors, focusing on three factors: advisor roles, essential qualities, and valuable behaviors. Machine learning models Decision Tree, Neural Networks, and K-Nearest Neighbors (KNN) were employed to categorize data and optimize advisor-student matching. The Decision Tree model demonstrated the highest efficiency, achieving 97.73% accuracy, 100.00% recall, 97.62% precision, and an F1-Score of 98.79%, making it the most effective for predicting advisor characteristics, expectations, and student satisfaction. This research provides a scalable solution for improving advisor-student matching, enhancing decision-making, and ultimately supporting the educational success of dental students.