Identification of sources of uncertainty in the evaluation of software systems quality

Today, software systems play a pivotal role in various aspects of life, with increasing complexity and diversity. However, along with this comes a growing number of challenges related to the quality and reliability of these systems. The need for assessing the quality of software products requires adherence to high accuracy and reliability through various methodological approaches. One of the main problems in this context is the insufficient substantiation of theoretical and methodological approaches in determining sources of uncertainty. This problem requires a comprehensive approach and systematization to address the tasks of evaluating the quality of modern software systems considering uncertainty. Significant contributions to the theoretical and practical aspects of development regarding the generalization and systematization of sources of uncertainty in the development and operation of modern software systems have been made by scholars such as C. Areces, R. Fervari, A. Saravia, F. Velázquez-Quesada, M. Bougeret, A. Pessoa, M. Poss, N. Boukhelifa, C. Johnson, K. Potter, L. Clarté, B. Loureiro, F. Krzakala, L. Zdeborova, D. Tsapetis, M. Shields, D. Giovanis, A. Olivier, et al. The study object is methodological approaches to determining sources of uncertainty. The study's purpose is to generalize the systematization of sources of uncertainty in the development and operation of modern software systems. To achieve the purpose, the following tasks are set and solved in the article: on a comprehensive level, the main sources of uncertainty in software systems are analyzed, such as changes in requirements, design flaws, unforeseen operating conditions, and other factors; a comprehensive model for optimizing the problem of formalizing uncertainty in software requirements is developed based on the application of machine learning methods and data analysis. In the process of this comprehensive research, methods of analysis, synthesis, generalization, and comparison are used. The author concludes that as a result of using methods' combination in the proposed development, it provides a more efficient and objective approach to managing the uncertainty of software requirements, compared to some existing approaches. This allows you to increase the reliability, quality, and efficiency of the developed software product.