«Systems and Means of Informatics»
Scientific journal
Volume 28, Issue 3, 2018
Content | About Authors
Abstract and Keywords.
- I. N. Sinitsyn
- V. I. Sinitsyn
Abstract: For nonlinear Volterra stochastic systems (VStS) with additive and parametric white noises (not obligatory Gaussian), exact and approximate methods for analytical modeling are considered. Analogies between mechanical and biological stochastic systems are discussed. Stability of stationary regular and irregular regimes in probabilistic first and second moments is studied. Special attention is paid to analytical modeling and stability analysis of stochastic processes in two-dimensional differential VStS. Some generalizations are considered.
Keywords: analytical modeling; Fokker-Plank-Kolmogorov equation; method of normal approximation (MNA); method of statistical linearization (MSL); normal (Gaussian) stochastic process; population dynamics; Pugachev equation; stochastic system (StS); Volterra stochastic systems (VStS)
- V. Yadrintsev
- K. Klyubina
- I. Tikhomirov
- A. Gershel'man
Abstract: The method of choosing a server solution for a digital platform of text search and analysis based on calculation of the total cost of ownership (TCO) is proposed. The method takes into account such parameters as the type of tasks solved by the digital platform, the required performance, the quality of service, etc. The proposed method is tried out on two modern server solutions in the face of various requirements and restrictions. The paper describes the hardware configurations, the features of the tasks to be solved and the parameters that were used in the calculation of TCO. The presented method of calculation can be widely applicable in solving practical problems of choosing hardware for digital platforms and data centers with known requirements and restrictions.
Keywords: digital economy; digital platforms; choosing of server solution; total cost of ownership, TCO; TCO calculation; text analysis; IBM Power8; Intel Xeon
- A. V. Borisov
- A. V. Bosov
- A. V. Ivanov
- A. I. Stefanovich
Abstract: The approach proposed earlier to evaluate the software performance based on simulation and statistical processing of experimental results is developed for the needs of staff management of an analytical information system (AIS).
To this purpose, staff responsibilities are classified to decompose the major AIS tasks solved by the corresponding application software. The paper presents methodical recommendations how to choose a set of timetable characteristics and observable metrics and collect related statistical information during the AIS testing and operation. The paper also contains the algorithms for calculation of the parameters of operations timetable and their statistical refinement. The methodology for assessing the performance of staff and the required number of staff is illustrated by simple examples.
Keywords: analytical information system; probability characteristics; statistical methods; factor analysis
- A. A. Kudryavtsev
- S. I. Palionnaia
- S. Ya. Shorgin
Abstract: The Bayesian approach to constructing models of the reliability theory is considered. Within this approach, the model is considered to be incomplete in a certain sense - it is assumed that the key parameters of the system are random variables with known a priori distributions. At some time points, the modifications are introduced to the system to improve reliability; however, each modification may either increase or reduce the reliability of the system. Thus, system's reliability characteristics depend on the ratio of the modification means' parameters of "efficiency" to the parameters of "defectiveness." Such relation can be called the "system's balance index." In this paper, the case of beta-polynomial a priori distributions is considered, where one of the parameters of the system has an a priori beta distribution and the density of the other parameter has the form of a polynomial. For various combinations of given a priori distributions, the formulas for calculating the probabilistic characteristics of the balance index are provided.
Keywords: Bayesian approach; modifiable information systems; reliability theory; polynomial densities; beta distribution; balance index
- A. K. Gorshenin
- V. Yu. Kuzmin
Abstract: Modeling and analysis of nonstationary data flows in real systems of various types can be effectively performed using finite local-scale normal mixtures. Approbation of the prediction methodology developed by the authors is carried out on the example of time-varied moments of the mixed probability model. Within this approach, values of the initial continuous time-series are replaced with the discrete ones and then modified samples are analyzed with a neural network. For short-term forecasting, the accuracy of more than 80% is demonstrated. Feedforward neural network is implemented using the Keras deep learning library, the TensorFlow framework, and the Python programming language.
Keywords: finite normal mixtures; moments; artificial neural network; forecasting; deep learning; data mining
Abstract: The article is devoted to the methods of evaluation of the waveform (or activation pattern) that occurs in the signals of the subject in response to an external stimulus during a neurophysiological experiment. The peculiarity of the mathematical model considered is the representation of noise in terms of finite normal mixtures. This approach is more consistent with the experimental data. Within the framework of the considered model, the analysis of different waveform estimators is carried out and their statistical properties are investigated. Several new estimators that utilize features of noise distribution were proposed.
The author proved higher efficiency of new estimators in terms of variance amount in relation to the method of moments which is widely used in studies. Despite the fact that the article focuses on the application of estimators to the magnetoencephalographic signals, the considered methods can be used in similar applications for any neurophysiological signals as well as in the most general setting - for bias estimation of a finite normal mixture.
Keywords: magnetoencephalography; signal processing; finite Gaussian mixtures; point estimators; efficient estimators; Fisher information; estimation of distribution bias
- K. R. Usmanova
- S. P. Kudiyarov
- R. V. Martyshkin
- A. A. Zamkovoy
- V. V. Strijov
Abstract: The authors analyze relationship and conformity between indicators in control system, monitoring of state, and accounting of railway cargo transportation. Macroeconomic time series that contain control actions, system state, and target criteria are considered. The authors suppose that control actions, state, and goal-setting are statistically related. Granger causality test is used to establish a relationship between time series. It is assumed that a pair of time series are related if the use of the history of one of the series improves the quality of the forecast of the other. The main goal of this analysis is to improve the quality of cargo transportation forecast. The computational experiment is carried out on data about cargo transportation, control actions, and set target criteria.
Keywords: time series; forecasting; Granger causality test; control system; target criteria
- A. A. Zatsarinny
- Yu. S. Ionenkov
- A. P. Suchkov
Abstract: The article is devoted to the effectiveness evaluation of the cloud technologies taking into account their specifics. The main characteristics and features of cloud technologies affecting their efficiency are considered. The generalized and partial performance indicators, taking into account the risks and vulnerabilities in the implementation of cloud technologies, are proposed. Mathematical models of cloud technologies efficiency indicators used in calculations are presented. A method of evaluating the effectiveness of cloud technologies, using a set of quantitative data characterizing all aspects of technology, combined with the use of expert assessments is developed. An example of evaluating the economic efficiency of cloud technologies implementation is given.
Keywords: cloud technologies; service model; efficiency; indicator; criterion; situation center
Abstract: The article deals with various scenarios of the interdepartmental information exchange system formation. They differ in the ways of the system core integration models formation. The first scenario assumes a minimal integration in the kernel of frequently encountered basic concepts. In the second case, the initial formation of the integration core ontology, approaching the full composition of the basic elements, is carried out. Integration of automated information systems (AIS) is carried out taking into account the existence of this core, but without changing it. The third scenario provides the formation of the core by attaching the ontologies of AIS as they integrate without creating an initial integration core.
The comparison of these scenarios by costs is carried out, the optimal scenario is defined. Two examples of national systems of interdepartmental exchange are considered.
Keywords: interdepartmental information interaction; integration scenarios; information data models; ontologies
Abstract: The paper deals with providing scientific services in the digital economy using a digital platform for scientific research. The authors propose a business model of a research services digital platform. The capabilities of a digital platform as a tool for managing scientific research in the digital economy are discussed. The main architectural solutions for the system providing scientific services of the digital platform are described.
Keywords: digital platform; cloud service; scientific service; information technology; information support activities; research management; unified information environment
Abstract: Virtualization increasingly penetrating present-day information and communication systems provides them with greater information security due to more reliable mutual isolation of multiple users and applications; it also provides system administrators' management efficiency and essential reduction of operating costs. All these advantages are achieved, however, at the expense of certain overheads that decrease the general performance of a virtualized system as compared with a physical machine. The overheads have a number of components, which depend on a level and degree of virtualization and are influenced by several factors, such as peculiarities of operating systems and hypervisors, or a specific character of the hardware virtualization support within server processors. Achieving satisfactory virtualization effectiveness is unfeasible without understanding the influencing factors and adequate estimation of the extent of influence regarding each of them.
Keywords: containerization; hardware virtualization support; machine level virtualization; operating system level virtualization; virtualization overheads
- A. Kolesnikov
- S. Listopad
- F. Maitakov
Abstract: Traditionally, the modeling of intelligent reasoning in hybrid intelligent systems is based on symbolic-logical models of knowledge. At the same time, experts, solving practical problems, operate mainly with visual knowledge. Visualization qualitatively changes the work of the subject of management, allowing one to recognize the problem situation at a glance and "see its solution" without logical inferences. The complexity of modeling visual reasoning in hybrid intelligent systems is due to the lack of formalized models of visual languages and high science-intensiveness of development of special environments for manipulating and processing visual models. The paper presents a formalized description of two basic levels of metalanguage for hybrid intelligent systems for managing electrical networks: (i) a conceptual and visual basis; and (ii) resources, actions, and properties.
Keywords: decision support; visual language; visual reasoning
- I. Kirikov
- A. Kolesnikov
- S. Listopad
Abstract: Traditionally, to solve problems in social and technical systems, methods of collective intelligence are used, which allow coping with the heterogeneity of information and the dynamics of situations. Decision making by a team of experts at a round table, a hybrid intelligent multiagent system, or a virtual heterogeneous team is a complex multistage negotiation and exchange of solutions process, for effective organization of which a relevant tool is needed.
A method for visual modeling of the synergy effect is proposed to visually assess the effectiveness of the team of experts and/or agents and to take control actions to improve it.
Keywords: expert team; synergy; hybrid intelligent multiagent system
Abstract: The stages of the process of discovering implicit knowledge about the translation correspondences of linguistic units in the original and the translation as well as the technology of goal-oriented creation of linguistic typologies based on the knowledge are considered. This is not a representation of conventional knowledge on the theory and practice of translation, explicated in reference books and dictionaries. An approach is proposed to identify gaps in the existing system of the knowledge about language and translation and informatics tools that ensure the goal-oriented filling of the identified gaps are considered. The description of the proposed approach is illustrated by the example of developing a typology of models for translating German modal verbs into Russian. The typology is created iteratively in the process of automated processing of parallel texts. In them, there are instances of the implication of subjective knowledge of translators that is not represented in the existing system of modern knowledge about language and translation. To provide explication, information technology is suggested that helps linguists process parallel texts and discover implied knowledge. The purpose of the article is to describe the stages of the goal-oriented discovery of knowledge implied in parallel texts.
Keywords: parallel texts; corpus linguistics; implicit knowledge; knowledge discovery; information technology; goal-oriented process; creation of typologies
- V. A. Minin
- I. M. Zatsman
- V. A. Havanskov
- S. K. Shubnikov
Abstract: The outcomes of the experiment on calculating the values of the intensity indicators of science-technology linkages are presented. The experiment was carried out using an eleven-year array of full-text descriptions of inventions.
The indicator evaluation of the process of knowledge transfer to the sphere of technology is an important element in the study of the linkages between science and technology, characterizing the processes of creating scientific foundations for developing advanced technologies. In the course of research conducted at the Institute of Informatics Problems of the Russian Academy of Sciences, a model, a methodology, and a testbed of an information system have been created for calculation of the values of the intensity indicators of science-technology linkages.
The main aim of this experiment is to compare the meanings of indicators of two types that characterize different aspects of the intensity of knowledge transfer to the sphere of technology. Comparison of the calculated values of these indicators is made and their meanings are specified.
Keywords: science-technology linkages; citation of publications in inventions; intensity of knowledge transfer process; indicator assessment; information technologies
Abstract: The review and analysis of existing methods of geodata generalization are presented and their specific features are revealed. The essence, determining factors, and content of the processes of spatial data generalization in the paradigms of analog and digital cartography are considered and compared. On this basis, the description of the concept of the property of variability of the generalization process is proposed, which, along with the properties of reversibility and alternative, is positioned as a starting point for creation of a unified methodology for reversible generalization of information objects.
Keywords: geodata; cartographic generalization; generalization of spatial data; digital cartography; geoinformatics
- I. M. Adamovich
- O. I. Volkov
Abstract: The article focuses on further development of T-parser, a system of automatic extraction of facts from historical texts, which is a component of the technology of historical and biographical research automation. The article analyses the defects of the current implementation, describes and substantiates the methods of their correction by excluding cycles from the grammar and its linear ordering. The description of the updated parsing algorithm and experimental verification of its effectiveness in comparison with the previous version carried out with real historical texts are adduced. The results of experiments which confirm the high efficiency of the updated algorithm and its applicability to the technology of historical and biographical research automation are described.
The technology is intended for a broad range of nonprofessional users, which is topical as the public interest to family history is increasing. The ways of further modification of the algorithm with the purpose of increasing facts extraction efficiency are outlined.
Keywords: facts extraction from texts; GLR-algorithm; pseudoorder; linear ordering; excluding cycles
Abstract: In the technology of designated payments, which is to be implemented in the digital twins environment, the rules of order execution and payment are rigidly linked. Each payment is certified by the special service of the bank- regulator. The state of the payment system and the paid orders is continuously modeled by their digital twins. The most complete implementation of the proposed technology should be done in online services of the banking system in the normalized economic mechanism. Such system includes bank-regulator, banks-providers, and corporate and personal electronic banks.
Keywords: payment system; technology of designated payments; digital twin; bank-regulator; bank-provider; corporate electronic bank; personal electronic bank; banking online service
Abstract: The problems of organizing the information retrieval in library catalogs both in the "historical" aspect (with reference to traditional card catalogs and printed catalogs) and in the modern aspect, conditioned by the development of information technologies, Internet, and universal search engines, are under consideration. The analysis is made and the shortcomings of the existing online catalogs of various scientific libraries of the country are revealed. A number of examples of incorrect query processing by catalogs of leading libraries of Russia are considered. The conclusion is made about the necessity of a new approach to the design of scientific libraries online catalogs. These catalogs should give the user all the information he/she needs without information noise. When creating software shells of modern online catalogs, on the one hand, it is necessary to move away from the traditional library ideas about catalogs, significantly limiting their search capabilities, and on the other hand - from the logic of the universal Internet search engines, giving the user thousands of percent of information noise.
Keywords: electronic catalogs; Internet; bibliographic descriptions; subject headings; network technologies; search requests; information retrieval
|
