Using electronic resources and neural technologies to develop competencies and predict academic performance in mathematics and computer science lessons

The research was aimed at developing and testing the system of predicting students’ academic performance based on the analysis of digital trace by means of neural networks in the conditions of secondary schools in Kyrgyzstan. The research methodology included creating electronic educational resources, developing a neural network model of forecasting and conducting a pedagogical experiment involving 120 students of 8-9 grades of secondary school No.42 in Osh city. The experimental group worked with the developed electronic resources including 27 interactive lessons, 250 test tasks and 30 practical works, while the control group followed the traditional teaching methodology. The monitoring system recorded 12 digital footprint parameters at 5-min intervals, which ensured high prediction accuracy. The developed system demonstrated a prediction accuracy of 88.9% for mathematics and 89.3% for computer science, which exceeded the performance of existing analogues by 4.9%. The introduction of the system contributed to the development of subject competences with the increase of the following indicators: algorithmisation – 34.1%, mathematical modelling – 32.9%, programming – 33.2%. In the experimental group, the average score in mathematics increased by 0.56 and in computer science by 0.63, while the quality of knowledge exceeded that of the control group by 14.4%. Advance prediction demonstrated 83.2% efficiency in identifying potential learning problems 2 weeks before the manifestation. The statistical significance of the results was confirmed by the high value of the Cohen effect (d = 0.85) and the correlation between systematic work with electronic resources and academic performance (r = 0.82). The developed methodology was adapted to the technical conditions of Kyrgyz schools and could be scaled up in educational institutions with a similar level of equipment in the presence of basic computer infrastructure and Internet access.