<![CDATA[This increase in adaptability is sought through the application of optimization techniques using the Gaussian Naive Bayes and K-Nearest Neighbor (KNN) methods. This research utilizes GridSearchCV to find optimal parameter configurations in both methods. The Gaussian Naive Bayes method will be used to analyze and classify student adaptability patterns based on historical data. In addition, the K-Nearest Neighbor (KNN) method will be used to utilize information from students who have similar characteristics to increase prediction accuracy. The main steps of this research involve collecting student adaptability data from online education sources, processing the data to obtain relevant features, and using GridSearchCV to find the best parameters in the Gaussian Naive Bayes and KNN models. By optimizing the prediction model using the GridSearchCV technique, this research is expected to make a significant contribution to improving the quality of online education, creating a more adaptive learning environment, and helping educational institutions in designing appropriate learning models. The Receiver Operating Characteristic (ROC) curve also showed a superior Area Under the Curve (AUC) score for KNN at 0.89, compared to GNB 0.81, confirming that the optimized KNN model offers significantly better sensitivity and specificity in predicting student adaptability levels in online education.]]>
Copyrights © 2025
