Predicting Student Academic Performance Based on Individual Cognitive Differences

Models were developed using the methods of decision trees and multiple regression analysis to predict the final academic performance of students for the entire period of study (four years) in the bachelor’s degree. The models take into account individual indicators of the cognitive potential of students, identified at the beginning of the first semester using modified computer methods for diagnosing cognitive styles. Approbation of the models was carried out on the data set obtained during the training process (StudCSP-DataSet-23). The results obtained confirm the scientific and practical significance of the conducted research in the development of learning technologies that take into account the individual cognitive differences of students in information processing.