<![CDATA[One of the challenges high school students face is the abundant availability of information about various campuses through different media, making it difficult to accurately predict their interest in a particular campus. Electroencephalogram (EEG) technology can read human brain activity, such as when students access information on a campus website. The Naive Bayes and K-Nearest Neighbor (KNN) methods can be employed to predict student interest in a campus based on EEG signals recorded while they browse the official campus website. Naive Bayes is known for achieving high accuracy with small datasets, whereas KNN excels at classifying noisy data. These two methods offer variables that can be directly compared. Classification using Naive Bayes and KNN achieved the highest accuracy score of 92%. The most appropriate algorithm is determined by evaluating performance using a confusion matrix. In this case study, Naive Bayes slightly outperformed KNN, as evidenced by precision, recall, and f1-score matrices. The Naive Bayes method resulted in an F1-score of 94%, compared to KNN’s 92%.]]>
Copyrights © 2024
