Artificial intelligence systems are increasingly being used in industrial applications, security and military contexts, disaster response complexes, policing and justice practices, finance, and healthcare systems. However, disruptions to these systems can have negative impacts on health, mortality, human rights, and asset values. The protection of such systems from various types of destructive influences is thus a relevant area of research. The vast majority of previously published works are aimed at reducing vulnerability to certain types of disturbances or implementing certain resilience properties. At the same time, the authors either do not consider the concept of resilience as such, or their understanding varies greatly. The aim of this study is to present a systematic approach to analyzing the resilience of artificial intelligence systems, along with an analysis of relevant scientific publications. Our methodology involves the formation of a set of resilience factors, organizing and defining taxonomic and ontological relationships for resilience factors of artificial intelligence systems, and analyzing relevant resilience solutions and challenges. This study analyzes the sources of threats and methods to ensure each resilience properties for artificial intelligence systems. As a result, the potential to create a resilient artificial intelligence system by configuring the architecture and learning scenarios is confirmed. The results can serve as a roadmap for establishing technical requirements for forthcoming artificial intelligence systems, as well as a framework for assessing the resilience of already developed artificial intelligence systems.

Publisher: Vilnius University Institute of Data Science and Digital Technologies, Journal: Informatics in Education, Title: Approach to Dynamic Assembling of Individualized Learning Paths, Authors: Vladimir LUBCHAK, Olena KUPENKO, Borys KUZIKOV , E-learning students are generally heterogeneous and have different capabilities knowledge base and needs. The aim of the Sumy State University (SSU) e-learning system project is to cater to these individual needs by assembling individual learning path. This paper shows current situation with e-learning in Ukraine, state-of-art of development of the adaptive e-learning systems and shows results of SSU research in this area. Nowadays the received solutions are different from the known analogues considering an expanded set of information about the features of a particular student's learning activities (19 indicators are analysed, including indicators of progress such as the level of knowledge and student individual features). The corresponding software solutions are being tested in the SSU e-learning environment.

The paper presents a method for classifying breast cancer histopathological whole-slide images using a deep learning model based on ResNet-50 and an Information-Extreme Learning (IET) classifier, which is crucial considering the increasing mortality rate from breast cancer year after year. The gold standard method of breast cancer diagnosis based on histopathological whole-slide images (HI) requires manual inspection of every region of the high-resolution image, which is time-consuming and error-prone. Computer-aided systems (CAD) powered by machine learning can speed up the work of a histopathologist and reduce the error rate of decision-making, thereby improving the accuracy and efficiency of the diagnostic process. Analysis of existing solutions has shown that deep neural networks, especially convolutional neural networks (CNNs), have the highest rate of functional efficiency in image classification tasks due to their ability to learn and extract relevant features from the input data automatically. However, the training stage demands a large amount of high-quality data, which is often scarce in histopathology, making it an open task to develop high-performance models with limited training image datasets. The proposed model was trained on the publicly available BreakHis dataset, achieving an 88.98% accuracy after only 20 training epochs, outperforming other methods. The results demonstrate that using a pre-trained CNN on a non-domain-specific dataset can yield accurate feature extraction. This shows promise for enhancing CAD systems, especially in data-limited scenarios. The model was compared with simpler models based on local binary patterns for feature detection and support vector machines for classification, highlighting its superior performance and potential for real-world application in breast cancer diagnosis.

The subject matter of the article is the application of supervised machine learning for the task of object class recognition. The goal is enhancing functional efficiency in information-extreme technology (IET) for object class recognition. The tasks to be solved are: to analyse possible ways of increasing the functional efficiency IET approach; implement an ensemble of models that include logistic regression for prioritizing recognition features and an IEI learning algorithm; compare the functional efficiency of the resulting ensemble of models on well-known dataset with classic approach and results of other researchers. The methods: The method is developed within the framework of the functional approach to modelling natural intelligence applied to the problem of object classification. The following results were obtained: The study tries to augment existing IET to support feature prioritization as part of the object class recognition algorithm. The classical information-extreme algorithm treats all input features equivalently important in forming the decisive rule. As a result, the object features with strong correlation are not prioritized by the algorithm's decisive mechanism – resulting in decreasing functional efficiency in exam mode. The proposed approach is solving this problem by applying a two-stage approach. In the first stage the multiclass logistic regression applied to the input training features vectors of objects to be classified – formed the normalized training matrix. To prevent overfitting of the logistic regression, a model the L2(ridge) regularization method was used. On the second stage, the information-extreme method as input takes the result of the first stage. The geometrical parameters of class containers and the control tolerances on the recognition features were considered as the optimization parameters. Conclusions. The proposed approach increases MNIST (Modified National Institute of Standards and Technology) dataset classification accuracy compared with the classic information-extreme method by 26,44%. The proposed approach has a 3.77% lower accuracy compared to neural-like approaches but uses fewer resources in the training phase and allows retraining the model, as well as expanding the dictionary of recognition classes without model retraining.

Web accessibility remains a critical aspect of ensuring equal opportunities for internet resource usage, especially for people with disabilities. The Web Content Accessibility Guidelines 2.5.3 criterion “Label in Name” requires that the accessible name of an interface component include text that is visually presented. Existing automated verification methods for this criterion are predominantly based on primitive string comparison, which does not account for semantic context. Objective: investigate the possibilities of using small language models with up to 1 billion parameters for automated semantic analysis of compliance with Web Content Accessibility Guidelines 2.5.3, as an alternative to resource-intensive large language models and limited algorithmic methods. Methodol ogy: the research involved creating synthetic datasets (7,200 English-language and 5,615 Ukrainian-language samples) and using real-world datasets (Top500 – 380 samples, UaUniv – 319). Sentence Bidirectional Encoder Representations from Transformers models were tested for computing semantic similarity, and fine-tuning of the google/electra-base-discriminator model was performed for 3-class classification of semantic relationships (“similar”, “unrelated”, “opposite”). Results: the trained model of 437 MB demonstrated high accuracy on syn thetic data (0.96) and sufficient accuracy on real datasets (Top500: 0.77, UaUniv: 0.73). The model effectively identifies all three classes of semantic relationships with an accuracy of 95.1 % for “opposite”, 92.7 % for “un related”, and 97.4 % for “similar” texts in the validation sample. Conclusions: the research confirmed the feasi bility of using small language models for automated verification of semantic compliance according to Web Content Accessibility Guidelines 2.5.3. The proposed approach provides acceptable classification accuracy with significantly lower computational costs compared to large language models, allowing for the integration of se mantic analysis into standard development and testing processes. Despite certain limitations, the developed so lution can significantly improve web accessibility testing.

In today’s digital world, website content accessibility for all users, including people with disabilities, is crucial. This paper examines the accessibility of web content on the websites of Ukrainian higher education institutions to assess their compliance with modern standards and requirements. The goal is to identify problematic aspects and develop strategies for improving the accessibility of educational web resources. The study covered data from the main pages of 372 higher education institution websites, whose addresses were obtained from the Unified State Electronic Database on Education (“USEDE”). The built-in accessibility enhancement tools used on these pages were analyzed. The paper also summarizes global experience in regulating web accessibility requirements. The analyzed regulatory documents include Web Content Accessibility Guidelines version 2.1 in their rules. Automated analysis tools WAVE and Axe were used to assess website compliance with the Guidelines. Results showed that the most common problems were insufficient highlighting of hyperlinks (characteristic of 84.8% of sites, Success criteria (SC) 2.4.4, 4.1.2), insufficient element contrast (84.6%, SC 1.4.3), lack of alternative text for images (42.9%; SC 1.1.1), and non-compliance with markup and semantics requirements. The conclusions present an analysis of the problems identified during the study, along with the authors’ recommendations for addressing them. The analysis results were systematized and published on a specially created web resource. The authors plan to conduct similar studies on an ongoing basis. It is also necessary to pay attention to problems identified when using automatic analysis tools. The Axe tool identified 11.875 cases of element contrast issues, accounting for 24% of the total number of detected problems. However, this figure is not final, as it does not account for the full range of possible color perception violations and overlooks the contrast of elements against backgrounds with non-uniform colors.