Algorithm for making recommendations in electronic educational environments based on stochastic Markov models

This article proposes a recommendation algorithm for educational resources in e-learning systems. The new approach uses Markov's model of evaluating the systems` content by casual users to form the parameters of the initial state, which characterizes a new user of the system as evaluations of the first resources (system content) to recommend interesting system elements for an active user. Thus, the problem of "cold-start" for the new users at the first phase of interaction with the system is solved. This problem is inherent in the system under development because the e-learning system includes a module for making recommendations, which allows it to refer to the class of recommendation-based automated systems. The new approach will combine the Markov process usage and the time factor to use them as a single data source for making recommendations. This approach will be based on the principle of access analysis of similar system users (the similarity is determined by comparing their profiles) in the same periods. An integral part of the created system is also usability. Therefore, at the design phase, it is necessary to think about the ergonomics of the recommendations in the educational system.

1. Drachsler H., Verbert K., Santos O. C., Manouselis N. Panorama of recommender systems to support learning. In: Ricci F., Rokach L., Shapira B. (eds) Recommender Systems Handbook, 2015. p. 421–451.

2. Ding Y., Li X. Time weight collaborative filtering. In Proceedings of the 14th ACM international conference on Information and knowledge management. 2005;1:485–492.

3. Yanaeva M.V., Sinchenko E.V. Study of recommenfation systems. Elektronnyy setevoy politematicheskiy zhurnal «Nauchnyye trudy KubGTU» = Electronic network poly-theme journal «Scientific Proceedings of KubGTU», 2017;2:104-114. Available from: https://ntk.kubstu.ru/data/mc/0039/1408.pdf (Accessed 12th december 2020). (In Russ.)

4. Vlasova E.Z., Balakireva E.V. Corporate environment of information and technological interaction of universities. Chelovek i obrazovaniye = Man and Education, 2011;3:45-48. (In Russ.)

5. Gosudarev I.B. E-learning: trends in model development and application experience. Известия Российского государственного педагогического университета им. А. И. Герцена = Proceedings of the Russian State Pedagogical University named after A. I. Herzen. 2013;162:162-166. (In Russ.)

6. Vlasova E.Z. Adaptive technologies as a means of optimizing the management of students' learning activity. Izvestiya Baltiyskoy gosudarstvennoy akademii rybopromyslovogo flota: psikhologo-pedagogicheskiye nauki = Proceedings of the Baltic State Academy of Fishing Fleet: Psychological and Pedagogical Sciences. 2011;4:6–15. (In Russ.)

7. Abtamov P.B., Len’shin A.V. Evaluation of the parameters of queuing systems, taking into account the aftereffect in the flows of serviced applications. Uspekhi sovremennoy radioelektroniki = The successes of modern radio electronics. 2013;9:45–48. (In Russ.)

8. Glazkova I.Y. Construction of a stochastic model of investment risk analysis. Ekonomicheskiy analiz: teoriya i praktika = Economic analysis: theory and practice. 2007;1(82). (In Russ.)

9. Matveev B.A. Spectral Theory of Risk. Vestnik YUUrGU. Seriya «Ekonomika i menedzhment» = Bulletin of the South Ural State University. Ser. Economics and Management. 2014;8(2):20–24. (In Russ.)

10. Popov A.A., Gul’tyaeva T. A., Uvarov V. E. Recognition, decoding and reconstruction of sequences with omissions described by a hidden Markov model with a discrete distribution of observations. Nauchnyy vestnik NGTU = Scientific Bulletin of the NSTU. 2017;1:99-119. (In Russ.)