For example,Бобцов

Intelligent adaptive testing system

Annotation

Modern learning is impossible without automated knowledge testing systems. At present, the most progressive are adaptive testing models in which the complexity of tasks varies depending on the correctness of the patient’s answers. This article describes the development of an intelligent adaptive testing system using a fuzzy mathematics device. An intelligent adaptive testing system has been developed; the module that implements the expert system uses the production base of the rules. The input parameters of testing are the percentage of correct responses, the degree of correctness of the response, the duration of the response, and the number of attempts. The output is a change in the current level of training of the student on the basis of which test questions of related complexity are selected. As a method of logical inference, the Mamdani method is used which consists of six operational actions: phazification — conversion of exact values of input variables into values of linguistic variables through belonging functions, this served as the basis for designing a fuzzy base of rules of the expert system; aggregation of sub-conditions — determination of the truth of conditions for each linguistic rule of the fuzzy inference system; activating sub-conclusions — finding the degree of truth of each of the sub-conclusions in the linguistic rule; accumulation of conclusions — finding the belonging function for each of the output linguistic variables; defazzification — finding a numerical value for each of the output linguistic variables. A developed intelligent adaptive testing system (ISAT) is presented that allows, based on the analysis of test results, to determine the current level of training of students, to adapt the material to the level of their training. This system allows you to dynamically present questions of appropriate complexity in real time. When using the developed intelligent adaptive testing system, students will be provided with questions of the appropriate level of complexity, this will allow building an individual learning trajectory. The introduction of a predefined system will ensure the implementation of a personalized approach for organizing the learning process; will increase the accuracy of assessing students’ knowledge. The use of the technology of fuzzy expert systems allows for automated, intelligent control of students’ knowledge.

Keywords

Постоянный URL

Articles in current issue

CONTENT © ITMO UNIVERSITY 1993-2024