MATHEMATICAL MODELS AND ALGORITHMS OF ADAPTIVE TESTING BASED ON NEURAL NETWORKS

Golovin Dmitry Aleksandrovich

The article discusses various approaches to the development of an intelligent system using mathematical models and algorithms for artificial intelligence (AI)-based adaptive testing. The focus is on the integration of mathematical methods and AI technologies to improve the accuracy and efficiency of testing – this includes the use of deep learning techniques to analyze test data and adapt tests in real time depending on the results and user behavior. Key adaptive testing algorithms, both classical and using neural networks and machine learning models, their application in different domains and their impact on results are analyzed. The research aims to optimize the testing process through dynamic adaptation and improved user experience to achieve more accurate and reliable results to achieve more accurate and reliable results to achieve more accurate and reliable results. The result of this study is an optimal block architecture of the intelligent system with a detailed description of all the existing blocks of the system as well as the processes of their interaction with each other. In addition, certain recommendations for further optimization of mathematical models and algorithms of adaptive testing based on artificial intelligence technologies are given.

Keywords:adaptive testing, artificial intelligence, mathematical models, intelligent system, machine learning, neural networks, information system, algorithms