Development of a Method for Assessing Learning Outcomes Through Automated Testing Management

Abstract

Testing knowledge is an important part of the pedagogical process, as it helps to improve the quality of professional knowledge, contributes to the effectiveness and efficiency of training. This article discusses the use of several approaches to building a fuzzy diagnostic expert system for knowledge testing. The aim of the research is a comprehensive optimisation of the process of mass test information processing in order to obtain reporting data, consideration of clear and fuzzy tests, drawing the analogy between clear and fuzzy tests, proposal and justification of different schemes of summing up the testingandproposal of one's own prototype of automated expert testing system which covers all stages of the testing process. Generally speaking, testing, as an algorithmic problem, can be represented as a problem of building an expert system to evaluate knowledge of testees. Such systems are used in solving many problems related to effective management, scientific research, and the like, where it is necessary to get the result with a limited amount of reliable information. It is established that the classical theory of tests has a number of controversial assumptions, and the results of its application have serious practical drawbacks. Therefore, the main principle of the approach in the work is the separation of the creation of a bank of test tasks and the process of compiling tests (bank of test material) and testing.