Development of the modern educational technologies caused by broad introduction of comput-er testing and development of distant forms of education does necessary revision of methods of an examination of pupils. In work it was shown, need transition to mathematical criteria, exami-nations of knowledge which are deprived of subjectivity. In article the review of the problems arising at realization of this task and are offered approaches for its decision. The greatest atten-tion is paid to discussion of a problem of objective transformation of rated estimates of the ex-pert on to the scale estimates of the student. In general, the discussion this question is was con-cluded that the solution to this problem lies in the creation of specialized intellectual systems. The basis for constructing intelligent system laid the mathematical model of self-organizing nonequilibrium dissipative system, which is a group of students. This article assumes that the dissipative system is provided by the constant influx of new test items of the expert and non-equilibrium – individual psychological characteristics of students in the group. As a result, the system must self-organize themselves into stable patterns. This patern will allow for, relying on large amounts of data, get a statistically significant assessment of student. To justify the pro-posed approach in the work presents the data of the statistical analysis of the results of testing a large sample of students (> 90). Conclusions from this statistical analysis allowed to develop intelligent system statistically significant examination of student performance. It is based on data clustering algorithm (k-mean) for the three key parameters. It is shown that this approach allows you to create of the dynamics and objective expertise evaluation.

expert system, data science, intelligent system, clustering, artificial intelligence

1. Dzharratano D., Raili G. Ekspertnye sistemy: printsipy razrabotki i programmirovanie. [Expert systems: principles and programming. 4 ed]. Moscow, Vil'yams, 2007. 1152 p. (in Russian).

2. Zhelnin M.E., Kudinov V.A., Belous E.S. The role and place of expert systems in education. Uchenye zapiski: elektronnyi nauchnyi zhurnal Kurskogo gosudarstvennogo universiteta [Scientific notes: electronic scientific journal of the Kursk state University]. 2012. no. 2 (22). pp. 1–5. (in Russian).

3. Baranova N.A. To the question of the application of expert systems in the continuous pedagogical education. Obrazovanie i nauka [Education and science]. 2008. no. 4 (52). pp. 24–28. (in Russian).

4. Andreev A.B., Moiseev B.M., Usachev Yu.E. The use of expert systems for the analysis of students ' knowledge in the environment of open education. Telekommunikatsiya i informatizatsiya obrazovaniya [Telecommunications and Informatization of education]. 2002. no. 2. pp. 36–54. (in Russian).

5. Berestneva O.G., Marukhina O.V. A computer system making decisions based on the results of expert assessment in the evaluation of the quality of education. Educational Technology & Society, 2002. no. 5 (3). pp. 216–230. (in Russian).

6. Dobre C., Xhafa F. Intelligent services for big data science. Future Generation Computer Systems. 2014. vol. 37. pp. 267-281.

7. Roiger R. J. Data mining: a tutorial-based primer. CRC Press, 2017.

8. Hovy E., Navigli R., Ponzetto S. P. Collaboratively built semi-structured content and Artificial Intelligence: The story so far. Artificial Intelligence. 2013. vol. 194. pp. 2-27.

9. Hutter F. et al. Algorithm runtime prediction: Methods & evaluation. Artificial Intelligence. 2014. vol. 206. pp. 79-111.