Abstract:

Pedagogical prediction in the context of digitalization of education plays one of the key roles in the successful adaptation of teachers to modern technological trends and allows to define possible changes and trends in education in the future. Also, it allows you to determine the directions of development of the educational system, identifies potential problems and propose strategies for solving them.

The purpose of pedagogical prediction is not only to define possible changes, but also to develop strategies and methods for adapting the educational system for the successful implementation of these changes. To build pedagogical prediction models, tools and methods for Educational Data Mining are used, which make it possible to create numerous prediction models. This paper describes the experience of training a model based on Seasonal Autoregressive Integrated Moving Average in order to assess the student’s learning outcomes on IT course for 2013-2022 and its further analysis depending on the obtained trends.

This study will use a dataset of the Results of 1332 bachelor students (“Pedagogical direction”) performance by using the final testing score data on the IT course for 2013-2022. Prediction of time series values was carried out in Python for the period 2023 to 2026.

Analysis of data from the study showed trends in the educational process for the course being studied, which allows us to draw the following conclusions:
1) adapt the content of the training course for the predicted period in the interval with a trend towards a decrease in the course learning outcomes;
2) the use of blended learning and the development of the resource using LMS Moodle for its implementation.

Thus, pedagogical prediction based on autoregressive integrated moving average model makes it possible to analyse the effectiveness of teaching for a given interval and, depending on the trends obtained, modify teaching methods and the content of educational programs to improve learning outcomes.

Keywords:

Pedagogical Prediction, Educational Data Mining, Python, Seasonal Autoregressive Integrated Moving Average Model.