Development of a Model for Predicting Student Academic Performance Based on Academic Indicators

The primary experiments proved the principal separability of objects based on a set of features characterizing students. In the course of further experiments, the development of classification models was performed to classify students into classes based on the selected features. In the course of the experiments, both types of models and algorithms for preprocessing of data sets were varied. From the resulting set of models, the best one was selected based on analyzing the performance on the test sample. After that, to visualize the work of the analytical system, a dashboard presenting the main academic indicators in different sections was developed.