Assessing the Reliability of Automated Knowledge Control Results

Recently the concepts of smart systems, smart education, smart engineering, smart medicine, smart city, smart universities, smart technology and smart communication were introduced in scientific methodology and practice for a modern society. All of these entities are the components of a new system - smart society system. These components must be equipped with corresponding connections that ensure their reliability, or, in other words, to have an appropriate smart infrastructure. One of the most important issues in this regard is a feedback and a control in smart system. This paper describes our approach to an application of statistical methods to automated knowledge control. The obtained research outcomes clearly show that if testing is used to test knowledge level, then a) it gives adequate outcomes of knowledge assessment only under repeated regular use of testing procedure and b) the final assessment should be obtained using the averaging procedure.