Parameters and Structure of Neural Network Databases for Assessment of Learning Outcomes - Pages 1638-1648

Eugeny Smirnov, Svetlana Dvoryatkina and Sergey Shcherbatykh

DOI: https://doi.org/10.6000/1929-4409.2020.09.188

Published: 14 December 2020

Abstract: The purpose of this study is to determine the methodology, develop a theory of construction, put into practice algorithmization and implement the functionality of a hybrid intelligent system for assessment of educational outcomes of trainees on the basis of the identified keyword parameters and structure of the artificial neural network using expert systems and fuzzy simulation; to develop a methodology for the construction of structural-logic, hierarchical, functional and fractal schemes for structuring databases of the didactic field of learning elements; to determine the content, structure of parameters and database components, selection criteria and the content of complexes of educational standards. The methodology of introducing intelligent systems into mathematical education is on the basis of the Hegelian triad: thesis (implementation of the coherence principle) – antithesis (implementation of principles of the fractality and historiogenesis) – synthesis (implementation of the principles of self-organization and reflection of the complex system inversion integrity). Requirements for the organization and construction of the artificial neural network for assessment of personal achievements on the basis of fuzzy simulation have been developed. In the direction of using elements of fractal geometry, the technological structures of clusters that constitute the basis of generalized structures have been developed. In particular, it is revealed that the didactic field of learning elements is equipped with a system of multi-level hierarchical databases of exercises, motivational-applied, research, practice-oriented tasks using expert systems and integration of mathematical, information, natural-science and humanities knowledge and procedures.

Keywords: Classification of learning outcomes, neural network, evaluation.

Download