Evaluation of attitudes of university students towards artificial intelligence using data mining methods

This study analyzes university students' attitudes towards artificial intelligence. Within the scope of the research, the data obtained from 1379 students through scale application were classified into three classes as "Insufficient", "Sufficient" and "Strongly Sufficient" according to their attitudes towards artificial intelligence. The data was classified by data mining methods. For this purpose, MLP, Decision Tree, KNN, XGBoost, Random Forest, CatBoost and SVM algorithms were used. The performance of the model was evaluated with a 5-fold cross-validation method. For each algorithm, basic metrics such as accuracy, precision, recall and F1 score were calculated and the classification performance was compared. According to the results, the highest F1-Score accuracy rate was 95.52% for the SVM algorithm. This was followed by CatBoost (93.66%), Random Forest (92.56%) and XGBoost (92.36%). The lowest success rates were observed in MLP (81.87%) and Decision Tree (82.72%) models. Confusion matrices revealed a tendency for frequent confusion with other classes, especially in the Strongly Sufficient class. The study concluded that advanced classification algorithms provide powerful and reliable tools for analyzing students' attitudes towards artificial intelligence. These findings may contribute to the development of educational policies and strategies for AI literacy.