Abstract:
The rapid integration of artificial intelligence (AI) into education has transformed how students learn, interact, and achieve academic goals. This study explores the impact of AI on university students’ academic performance, focusing on its influence on productivity, mental health, and retention. Using a survey dataset comprising 360 responses, machine learning models, including Random Forest, were applied to predict and analyze variables such as the effectiveness of AI tools in mental health support, sustained positive impact on academic performance, and productivity enhancement. Data preprocessing techniques like handling missing values, outlier removal, and encoding ensured data integrity, while the SMOTE algorithm addressed class imbalances. The Random Forest model achieved accuracies of 81.75%, 85.47%, and 87.50% for key variables, with macro average F1-scores highlighting the balanced performance across classes. These findings underscore the potential of AI in fostering deeper learning, enhancing mental well-being, and driving academic success. The research also examines the ethical and social implications of AI in education, emphasizing the need for responsible and inclusive integration strategies. By bridging existing gaps in understanding AI’s role, this study provides actionable insights for educators, policymakers, and stakeholders to harness AI’s transformative capabilities effectively. Future work includes expanding the dataset, exploring additional variables, and benchmarking against advanced models to deepen understanding and maximize AI’s educational potential.
