Dement'ev Evgeniy Aleksandrovich, Student, Penza State Technological University (1a/11 Gagarina street/Baydukova avenue, Penza, Russia), Bfh58@mail.ru
Kan Vitaliy Vyacheslavovich, Postgraduate student, Penza State Technological University (1a/11 Gagarina street/Baydukova avenue, Penza, Russia), firstname.lastname@example.org
Svistunov Boris L'vovich, Doctor of engineering sciences, professor, sub-department of physics, Penza State Technological University (1a/11 Gagarina street/Baydukova avenue, Penza, Russia), email@example.com
Background. The subject of the research is the synthesis of algorithms for assessing the quality of training of university students as the degree of formation of prescribed competences. The purpose of this work is to describe and discuss the proposed algorithm for identifying the actual status of the learner as a monitoring object and the procedure for classifying the detected state to a given a priori class of states using the Bayesian decision rule.
Materials and methods. The research is based on a systematic approach, namely, the study of a learner (monitoring object) as a complex stochastic multiply connected system. To assess the state of the system, recognition methods are used with recognition results processing according to the Bayesian classifier.
Results. An algorithm for recognizing the state of a monitoring object and a method for forming a decisive rule for assigning the state of an object to a certain class of states corresponding to the adequacy of the object to the requirements of regulatory documents are proposed and justified. The risks of recognition and classification procedures are assessed.
Conclusions. An approach is proposed for assessing the quality of students' training, based on the technique for determining the state of technical systems adapted for the researched subject area, taking into account the complexity of the behavior of the researched objects. On the basis of this approach, a simple, universal, adapted for practical application algorithm is proposed, relatively easily implemented using standard software and hardware.
1. Belashov P. D. XXI vek: itogi proshlogo i problemy nastoyashchego [XxI century: results of the past and problems of the present]. 2015, vol. 2, no. 6 (28), pp. 79–84.
2. Gusyatnikov V. N., Sokolova O. Yu., Kayukova I. B. Vestnik Saratovskogo gosudarstvennogo sotsial'no-ekonomicheskogo universiteta [Bulleting of Saratov Socio-Economic University]. 2009, no. 4, pp. 197–200.
3. Bespal'ko V. P. Prirodosoobraznaya pedagogika [Natural pedagogy]. Moscow: Narodnoe obrazovanie, 2008, 512 p.
4. Rastrigin L. A., Erenshteyn M. Kh. Adaptivnoe obuchenie s model'yu obuchaemogo [Adaptive learning with a student model]. Riga: Zinatne, 1986, 160 p.
5. Kurilova O. L., Smagin A. A., Lipatova S. V. Uchenye zapiski Ul'yanovskogo gosuniversiteta. Ser.: Matematika i informatsionnye tekhnologii [Bulletin of Ulyanovsk State University]. 2012, no. 1 (4), pp. 246–257.
6. Tsipkin Ya. Z. Osnovy teorii obuchayushchikh sistem [Fundamentals of the theory of learning systems]. Moscow: Nauka, 1970, 232 p.
7. Ivakhnenko A. G. Samoobuchayushchiesya sistemy raspoznavaniya i avtomaticheskogo regulirovaniya [Self-learning systems for recognition and automatic regulation]. Kiev: Naukova dumka, 1969, 134 p.
8. Tu Dzh., Gonsales R. Printsipy raspoznavaniya obrazov [Principles of pattern recognition]. Moscow: Mir, 1980, 411 p.