Eureka! Institutional Repository

The remarkable progress of both machine learning and data mining in recent years has made a decisive contribution to the field of education through the production of new knowledge, which contributes to decision-making and the improvement of the educational process. In the context of this research, various machine learning techniques will be applied with the goal of predicting students' academic performance. The research is divided into two periods, before and after the Covid-19 period, and concerns secondary school students as well as university students. In the first part, the technique of linear regression and neural networks will be applied in order to predict the academic performance of male and female students in the Mathematics course. The models will then be evaluated through different performance indicators to select the one that performs best. In the second part, the interest will be focused on the Covid-19 period, during which university students answered various questions in order to collect the second set of data that will be used. The prediction in this part will be about the average score of the final period and will be achieved by applying five different machine learning algorithms. As in the first experiment, in this one it is crucial to evaluate the algorithms in order to arrive at the optimal one based on important indicators such as accuracy etc. Educational data mining and analysis can be considered extremely useful tools for the development of the educational process. With their proper utilization, it becomes possible to identify possible factors that contribute to the academic performance of students and to prevent various problems that plague the field of education.