Vol 25, No 2 (2019)

The work is devoted to the problem of synthesis of multicomponent interoperable structures (MIS), which belongs to the class of open information systems structurally-parametric synthesis problems. A generalized method for synthesizing MIS based on the standardized components ontological representation, their interfaces and formed multicomponent structures is proposed. It is shown that the synthesis of MIS can be reduced to the task of managing interacting processes for the organization of distributed processing of ontology as a shared, distributed resource. A system of events is proposed, presented in the form of a nondeterministic finite state machine, which allows one to formalize and implement algorithms for controlling the multiprocess synthesis system for MIS.

The paper deals with the task of formalizing the description of a class of images of a person’s face. The research is aimed at finding methods that increase the accuracy of the description for recognition problems. The existing methods of description are briefly considered, such as nearest neighbor methods, separating hypersurface methods, support hypersurface methods, and probability density function methods. A technique is proposed that makes it possible to reduce the total errors of the first and second kind when describing image classes based on the probability density function. The problem of reconstructing the probability density function is also considered. Based on the experiments carried out, the advantage of the nonparametric Rosenblatt—Parzen method was revealed in comparison with the parametric approach, and a decision was made to apply this method to restore the density function based on experimental data. The recovery of probability density functions based on the Rosenblatt—Parzen method is interrelated with the search for a vector of optimal blur coefficients for each function. To solve this problem, it was proposed to use the gradientless Nelder—Mead method known as the simplex method. As a result:

• The description of the realization of the method of formalized description of the image class based on the reconstruction of the probability density function based on experimental data with the help of the non-parametric Nelder—Mead method was proposed
• Testing of this method was performed in comparison with the classical parametric reconstruction of the probability density function and the advantage of the proposed approach was revealed.

One of the major obstacles of expanding research in the field of personified medicine is the lack of technology for storing, processing, analyzing and verifying of biomedical information. The paper proposed the concept of biomedical data and knowledge warehouse, which combines information from different sources of various nature. The architectural feature is the integration of primary clinical and statistical data, results of their processing and analysis, combination of the latter with known medical knowledge on diagnostics and disease treatment and their use in clinical medicine. Heterogeneous warehouse has a component distributed architecture and combines information (ontologies, data and knowledge bases) and software (system, specialized and applied) units. Ontologies are divided into three layers: infrastructure (meant for infrastructure support of the warehouse), information (support formation of data and knowledge components) and software (these ontologies set structures for software components). Data is divided into primary (received form outer sources — case records, formalized data of the medical organizations, medical statistics, monitoring logs) and secondary (acquired while data analysis by statistical methods, machine learning and artificial intelligence). Knowledge bases describe diagnostics, cures and pharmacotherapy, treatment assignment. Several information resources and software components are described (medical terminology and observations glossary, export/import tool, clinical data preprocessing tool, statistics data preprocessing tool, specialized shell for decision support system development). To date, many software and information components of the warehouse have been developed and are being used, work continues on the creation of new components and the improvement of existing ones.

A complex queuing system is considered, as a complex of interacting tools that solve a common problem in accordance with the objective function and with the required values of the service parameters. The objective function displays the specific purpose of a complex system with the required values of the main service parameters in accordance with the terms of reference for its development. Concerning the values of these parameters, the condition of their maximization or minimization can be specified. Modernization of the system counteracts two main processes of degradation: "physical" aging and moral aging. However, during the operation of a complex system, structural defects of facilities and their controls, as well as software errors, may be revealed. In the general case, complex systems have a centralized management body and form a centralized network, the nodes of which display individual objects. The concept of object rating is introduced, which reflects the degree of its participation in providing high quality of a complex system. For each type of objects, using the suggestions of experts, a rating system is built on the logical neural network. With its help, the monitoring results determine the current rating of the object. If the rating has fallen below acceptable, the object claims to be repaired, replaced, upgraded, etc. In the conditions of financial and technological constraints, the optimal decision on many objects is made on the basis of the scale of their importance.