22 August 2023 in Technology audit and production reserves
The scientific task, which is solved in the research, is the analysis of knowledge representation methods in intelligent decision-making support systems. The problem is explained by the fact that the form of knowledge representation significantly affects the characteristics and properties of the system. In order to operate all kinds of knowledge from the real world with the help of a computer, it is necessary to carry out their simulation. In such cases, it is necessary to distinguish knowledge intended for processing by computational devices from knowledge used by humans. In addition, with a large amount of knowledge, it is desirable to simplify the sequential management of individual elements of knowledge. A homogeneous representation leads to a simplification of the logic management mechanism and a simplification of knowledge management. The research is aimed at the analysis of knowledge representation methods in intelligent decision-making support systems. Currently, many models of knowledge representation have been developed. The main models include: logical models; frame model; network models (or semantic networks); production models. Therefore, the object of research is the intelligent decision-making support system. The subject of research is an intelligent decision-making support system. The following is set: – the methods (models, approaches) presented in the research for presenting knowledge in intelligent decision-making support systems in a canonical form are not advisable to use for a number of objective reasons given in subsection 3.1 of the research; – it is necessary to develop new (improvement of existing) representations of knowledge in intelligent decision-making support systems, which will have the advantages of these approaches without their disadvantages. Further improvement of these approaches to reduce the number of shortcomings and limitations of their application should be considered as the direction of further research.
19 December 2025 in Problems of Informatization and Management
Efficient resource distribution in heterogeneous distributed computing systems requires intelligent node selection mechanisms capable of adapting to dynamic system conditions. This research presents a comparative evaluation of Convolutional Neural Network (CNN) and Multilayer Perceptron (MLP) architectures for binary node suitability classification in distributed task scheduling environments. The study employs five synthetic datasets ranging from 100 to 2000 instances, with each node characterized by four critical attributes: performance, security level, baud rate, and reliability. Experimental results demonstrate that the MLP architecture achieves validation accuracy between 91% and 98.8% with high consistency across dataset sizes, while the CNN architecture shows fluctuating performance ranging from 85% to 94%. The key finding reveals that MLP architectures outperform CNNs for tabular node data due to better alignment with unstructured attribute relationships, as fully connected layers naturally handle unordered features without imposing spatial locality assumptions. This empirical analysis provides practical guidance for architecture selection in AI-based resource schedulers, demonstrating that simpler architectures can outperform more complex ones when appropriately matched to problem structure. The findings contribute evidence-based recommendations for distributed systems practitioners implementing neural network-based scheduling solutions.