Software & Systems

Articles of journal № 1 at 2018 year.

Order result by: Public date | Title | Authors |

11. Evaluation of a degree of impact of access control means on information system efficiency [№1 за 2018 год]
Author: Drobotun E.B.
Visitors: 5440
There are requirements to effective performance of protective functions by protection systems against information and technical impacts and computer attacks. In addition, such systems are required to have minimal influence on protected system functioning and, in particular, on its performance. Ideally, a security system (or means) should not have any effect on the performance of a protected information system. At the same time, it should be noted that it is practically impossible to create a protection system (including the system of access control and delimitation) that would not have any negative effect on a protected object. Therefore, one of the criteria for choosing a system (or means) for access control to protect any information and computing system should be the criterion of the minimum effect of the chosen access control system on the performance of a protected information system. The article proposes the evaluation technique of information system performance degradation at introduction of access control means to it, which are one of the most important components of an information security system for information systems. The paper also describes software that helps implementing the offered technique.

12. Software to solve the problems of optimizing transport objects placement based on the algorithm of projection clustering [№1 за 2018 год]
Author: B.A. Esipov
Visitors: 5318
The choice of the logistics centers optimal placing is an important task for all types of transportations. Solving it for real problems of a region or country leads to mathematical algorithms of high complexity. The paper proposes a mathematical model and a new method for solving the problem of logistics centers optimal placement of a two-tier transportation network based on the mathematical apparatus of a cluster analysis. As geoinformation parameters of supplier plants, as well as a road topology, stations or other objects of transport infrastructure are given, there is the problem of an optimal choice of logistics centers. For example, logistics centers for a railways network can be container points. The optimization criterion is minimization of the total traffic volume in ton-kilometers from production to container points. For this purpose, the model of partitioning objects into clusters is used as an optimization mathematical model. The required clusters are subsets of production points with their own centers (container points). Since cluster centers must be located at railway stations, the article suggests a new clustering algorithm with a projection. The paper investigates the possibilities of such clustering algorithm called k-means pro. It also considers the methodology of selecting the number of centers (clusters) by a generalized economic indicator of transportation costs and costs for creating logistics centers. Based on the created software, there are some examples of calculations for enterprises and railways of the Volga Federal District.

13. Cost estimation in software projects design using the mountain algorithm [№1 за 2018 год]
Authors: Zubkova, T.M., E.N. Natochaya
Visitors: 6156
Cost estimation in software development allows reducing the probability of making incorrect or not making necessary management decisions, the probability of obtaining unplanned results of software projects. The article performs the analysis of basic approaches to the problem of software projects cost estimation: the linear approach, the method of function points and using empirical data. The paper proposes the technique of cost estimation in software projects design based on the mountain algorithm. The paper describes mathematical support and software to solve the problem. The task is implemented according to the application composition model and the early design model. To estimate software project costs it is offered to record the price of programs according to their group of complexity. The paper considers three groups of software products complexity: simple (design takes up to 4 months), average (design takes from 4 months to one year), high (design exceeds a year). The paper explains the use of the mountain clustering method for combining software products in groups based on a similarity of signs. Based on the offered method, the software system of software projects cost estimation is developed in order to increase objectivity and efficiency of the decisions made by developers. Input data of the software system are a structure of an enterprise, a specification, a table of estimates of software product elements, the Boehm table. The algorithm of a mountain clustering acts as controlling influence. The resources supporting execution of software system functions are hardware and software of the company. At the system output, the report on costs of software product design is created.

14. Multi-agent approximation of social media users’ behavior [№1 за 2018 год]
Authors: Ivaschenko A.V., A.A. Khorina, A.R. Diyazitdinova
Visitors: 5899
Nowadays social media are becoming one of the major Big Data providers for analyzing users’ behavior. At the same time each Internet user activity appears to be an interesting and complex object for formalization and study. One person can be presented in several social networks using various avatars, have several parallel interests, have own specific strategy to publish and process data in the form of posts, messages and comments and interact with the others by means of data exchange. Considering the specifics of social media users activity, who are relatively independent and self-organized, this paper proposes a model of such a behavior implementing the multi-agent approach. The paper describes the problem, solution architecture and technical details of its implementation supported by the results of simulation and real data analysis for a number of popular social networks. It is the first time when it is proposed to use the definition of “behavior multi-agent approximation”. The term introduces an approximate description of real use cases of the integrated information space users via multi-agent simulation. Within the framework of the solution is it proposed to represent the interest of social media users as a combination of a context and focus. Such approach allows formalizing deviations of user interest under the pressure of incoming events considering a combination of time and human factors. The proposed approach gives a possibility to describe and implement the patterns of typical Internet social media users behavior for simulation and analysis in order to detect trends and deviations.

15. The open cube concept for analysing social media big data [№1 за 2018 год]
Authors: Ivaschenko A.V., N.M. Shlychkova, V.A. Isayko, P.V. Sitniko
Visitors: 5680
Modern social media can be treated as an important source of Big Data describing users’ behavior during informational exchange. Understanding the basic trends at this level can help to determine the main behavior features, identify informational influence and analyze deviations of users’ interest to various informational objects and events, etc. On a practical level, this kind of analytics allows implementing an effective context-based advertising, increasing the efficiency of social networks functionality and solving various problems of information security. Analysis of Big Data that characterize social media users’ interaction appear to become a complex technical problem. The reasons are: it is required to integrate with several social networks for data import, to associate independent profiles of the same users at different networks, to correlate the facts of their interaction with real events and determine basic trends and deviations. To solve the problem the authors propose to implement a technology of “open cube” based on an orthogonal indicators system describing the data change dynamics in time depending on different factors. It is proposed to analyze distribution of incoming user interaction events, dynamics of their interest evolution in time, reaction to incoming events, etc. using cross-correlation analysis of time series using interval-based functions. The paper describes the basic problems of Big Data analysis in social media, the proposed abstract model of the open cube and the data analysis algorithm that allows identification of users’ activity at social media. The described model and its implementation were tested using a typical data set derived from a number of social networks. In addition to a real regular data set of social media users’ negotiation there was also introduced a series of messages generated by a bot, which was successfully identified using the proposed approach.

16. Multilingual thesaurus development framework [№1 за 2018 год]
Authors: A.V. Kirillovich, A.M. Bashirov , A.R. Gatiatullin
Visitors: 8274
The paper describes a new tool for creating a bilingual Russian-Tatar thesaurus. The paper lists the requirements for the implementation of the tool: support for selected formalism of thesaurus knowledge representation, support for multilingual thesauri, teamwork support; publishing a thesaurus in the Linking Open Data cloud (LOD); cross-platform and free licensing. The paper gives a detailed analysis of the existing editors of general purpose ontologies and specialized thesaurus editors based on SKOS, WordNet and RuThes models, which include: Protégé/WebProtégé, VocBench, PoolParty, iQvoc, SKOSEd, OMWEdit, WordNetLoom, GernEdiT, DEBVisDic, WordNet Editor, YARN and RuThes editors. It also describes the advantages and disadvantages of these editors and, therefore, it explains the choice of the RuThes model, and the reasons of developing an own tool that meets all the initial requirements. The new tool for developing a bilingual thesaurus is implemented as a web application with support for teamwork and publishing the thesaurus data in LOD. The paper describes the user interface (UI) both for the end-user, and the database editor modes. It is stressed that the software can be used not only for developing a Russian-Tatar thesaurus, but also for other languages. The goals for the further development are: UI refining, including advanced graphical visualization; integration with electronic cases and additional dictionaries; LOD publication module finalization; a universal tool for Lemon ontology-based lexical resources development; individual lexical units detailed description module development.

17. A multy-agent simulation of infecting objects dissemination and interaction [№1 за 2018 год]
Author: K.E. Klimentev
Visitors: 9546
The article discusses the main approaches to multi-agent simulation of infection processes typical of technical networks and wildlife. The examples of such processes are: the epidemics of pathogenic organisms among living creatures; dissemination of computer viruses and worms; expansion of forest fires; message transfer in technical and social networks, etc. The paper briefly describes traditional approaches to solving this problem and related problems. The article describes a general model of "universe". It also introduces the basic sets of objects and their attributes, considers relations between them typical of simulated processes and phenomena. It analyzes space graph models, such as: a full graph, a scale-free graph, a geometric graph and various grids. The paper describes basic states of agents. It also considers different models of state transitions, including SI, SIS, SIR, SEIR, SIDR, etc. It discusses various methods for target scanning, such as: sequential scan, random scan, list scan, counterattack and so on. It describes the method for setting the temporal characteristics of individual agents. One of the important features that distinguishes this development from its analogues is the possibility of modeling epidemics in changing universes. For example, it is possible to change the topology of graphs, object numerical attributes and agent behavior algorithms during the run. The use of such models would allow obtaining results that are more adequate. The paper also examines the architecture of a modeling system that supports of aspects discussed above. There is an example of a particular software implementation.

18. Study of the influence of initialization algorithms of Volterry network weight coefficients on forecast problem solving [№1 за 2018 год]
Authors: I.V. Lyozina, K.V. Saraeva
Visitors: 10189
The article describes an automated system designed by the authors for forecasting oil market quotes by Volterry neural network. The system runs three algorithms for weight coefficients initialization: the annihilation simulation algorithm, the evolutionary algorithm and random initialization. A user is able to upload samples, set learning parameters, select a learning algorithm for weight coefficients initialization, define their parameters and look through forecasting results. Moreover, a user can save parameters of a neural network to a XML file to provide an ability to perform forecasting by using preset parameters of a trained network. While testing a user can open a network saved earlier or work with a current one, load a sample file and then look at the results. All results are kept in .xlsx files. The developed software can be used for forecasting a market oil price. Standard deviation is an evaluation criterion of the designed application. To make a testing experiment the authors selected Brent and WTA oil prices during the period from December 31, 2005 till December 31, 2016. The article shows the results of researching a dependency of forecasting quality according to the selected algorithms. The results show that a random initialization (learning coefficient 0.01, a number of neurons L=3, a number of layers K=3) produces a bigger error than the annihilation simulation algorithm or the evolutionary algorithm. As a result, using optimization algorithms for initialization of weight coefficients provides more accurate forecasting result than a random initialization.

19. An enterprise as a purposeful system [№1 за 2018 год]
Author: Yu.M. Lisetskiy
Visitors: 6796
The article considers an enterprise as a purposeful system. It gives general information on an organization and a scientific apparatus describing its building and functioning. The paper describes a hierarchical structure of the enterprise as an organization with the main purpose in an economic sphere, as well as its functional model. The author analyzes characteristic features of the enterprise in the context of general principles of system building and functioning. There are some approaches and methods to studying an enterprise as a complex production system based on the system approach. The system approach enables complex estimation of any production and economic activity as well as the management system according to specific characteristics. This approach also enables analysis of a specific situation within a separate system, identification of input, process and output issues. The system approach enables best possible organization of a decision making process at all enterprise management system levels.

20. Architecture and algorithms of the intellectual support system for household sector functioning control process [№1 за 2018 год]
Authors: E.A. Makarova, E.Sh. Zakieva, R.R. Valitov, E.S. Gizdatullina
Visitors: 5767
The paper presents the structure of the intellectual support system to manage a household sector functioning process. The sector interacts with other sectors incorporated within the multi-sectoral macroeconomic system in general. As a controlled object the paper considers functioning of a multi-sectoral macroeconomic system that includes four sectors (real sector, financial institutions and general government sectors, as well as household sector decomposed to regional clusters) and three macroeconomic markets (amenities market, labour market, and money market). The management system structure consists of three management levels. The first management level is based on the feedback control approach and includes computation algorithms of balanced planned rates of population resource earnings and expenditures in the built regional clusters. It is designed for adjusting behaviour of economic sectors during implementation of primary functional processes. The second management level is based on the adaptation principle and is designed for adjusting planned rates of resource expenditures by economical sectors considering a household sector cluster decomposition based on the information on volumes of accumulated resources of the economical sectors, including regional clusters of the household sector; as well as information on the current level of minimum subsistence level, price levels on the amenities markets, labour market, and money market. The third management level corresponds with the level of governmental control over economy in income policy. It is based on the situational management principle and considers controlling actions in fiscal policy and income policy correspondingly. In order to implement controlling actions at this level, intellectual data analysis algorithms and models should be used, as well as connectionist and neuro-fuzzy models and techniques. The structure of the intellectual support system to manage functioning process of the household sector can be used in the software development for decision making support and simulation MMES system including formation of its major functional components and teir interconnections.

◄ ← Preview | 1 | 2 | 3 | 4 | Next → ►