THE METHOD OF DEVELOPING AN INTELLIGENT AUTOMATED SUBSYSTEM FOR ASSESSING THE LEVEL OF COMPETENCE FORMATION

Abstract

In the article the question of an estimation of reliability of level of formation of the competences on the basis of indistinct production cognitive maps is considered. The paper considers the issue of assessing the reliability of competence level based on fuzzy production cognitive maps. The automation of the process of assessing students’ competence level remains one of the urgent tasks in the field of informatization of education. However, the organization process of measuring and assessing the competence level, which demands the usage of information technologies, is currently not fully determined. This is due, inter alia, to the lack of recommendations for expanding the functionality of the existing informational systems used in higher education institutions for monitoring their educational process in the context of changes in the functioning environment. To monitor the forming of competencies, it is important to develop a subsystem for assessing the competence level as a part of the information system of educational organization. In modeling the process of assessing the level of development of students' competencies, weakly structured non-formalized information is used, therefore, it is advisable to use cognitive technologies when building models. The analysis of existing methods for formalizing knowledge and experience of experts allowed to determine that when developing a subsystem for assessing the level of competence formation, it is advisable to combine (hybridization) technologies that use, depending on the type of tasks, the positive properties of production, fuzzy and neural network models that complement the methods of classical expert assessment. However, the insufficiency of the methodological substantiation of the development of models and algorithms of the subsystem for assessing the level of competence development significantly reduces the reliability of the estimates obtained. The report proposes a methodology for developing a subsystem for assessing the level of competence formation, which allows creating information-mathematical support for such subsystems based on competence models, algorithms for their structural and parametric identification, and databases that help to assess the competence level. This method describes a system of competence level assessment based on reliability criterion. Thus, the proposed methodology for developing a cognitive subsystem improves the accuracy of assesing the level of competence development based on the creation of its information and mathematical software using fuzzy production cognitive competence models, neuro-fuzzy algorithms and the fuzzy Mamdani fuzzy inference rule base for their structural and parametric identification. At the end of the work, it is also proposed to consider a number of sources devoted to the formation of students' competences in subject areas [4-34], which will be considered in the further works of the authors.

Keywords

• Competence
• competence approach
• cognitive maps
• cognitive modeling
• intellectual system
• reliability