ISSN 0236-235X (P)
ISSN 2311-2735 (E)

Bookmark

Next issue

2
Publication date:
16 June 2019
-->

Articles of journal № 3 at 2016 year.

Order result by:
Public date | Title | Authors

1. About table inheritance effectiveness in PostgreSQL DBMS [№3 за 2016 год]
Author: Sorokin V.E.
Visitors: 7049
One of the consequences of extending the use of object-oriented programming was the addition of a relational DBMS the object features. On cross-platform open source object-relational DBMS PostgreSQL hopes in import substitution infrastructure software. It combines classic relational database ACID principles with complex data types and table inheritance support that are in demand in the simulation objects in the database. When working with the built-in complex type jsonb PostgreSQL is reached object-oriented DBMS performance. For the case of the preferred use of the base and composite data types for modeling objects in the database table inheritance is implemented. The article presents a comparative analysis of this mechanism with the most known methods of object-relational mapping. PostgreSQL developers planned but not yet implemented support single indexing for physical tables of one inheritance hierarchy is marked by them as the main factor restricting the use of inheritance tables due to the lack of support in unique and referential integrity and reduce the performance of the queries. The proposed in addition to existing table inheritance mechanism creating one common table for the hierarchy inherited tables allows you to ensure data integrity. Performed experiments on the execution different queries the properties of the objects modeled in the DB by the classical models of object-relational mapping and the proposed amended inheritance tables confirmed the high effectiveness of the latter, which allows to recommend it for widespread use, especially with the prospect of a single indexing hierarchy of inherited tables.

2. Probabilistic inference in weakly formalized knowledge bases [№3 за 2016 год]
Authors: Poleschuk E.A., Platonov A.V.
Visitors: 4808
The article introduces the process of probabilistic inference in weakly formalized knowledge bases. Semantic network is chosen as a graphical model of knowledge representation due to a convenient representation of automatically extracted data as a graph with links. The paper also contains a comparison of widely used production model approach with the proposed one. The authors describe main disadvantages of a production model approach that should be considered for developing such knowledge extraction systems. The aim of the work is new knowledge extraction from an automatically built knowledge base. Usually logical inference is used to achieve this aim in graphical models. In our case a domain model as well as a process of knowledge building (in particular by automatic or semi-automatic methods) restrict logical inference mechanism, so this algorithm is forced to work in conditions of uncertainty. Thus, standard logical inference algorithms provided for such model become irrelevant. The article proposes using probabilistic inference for the task and consequently using probabilistic inference programming language. The paper contains a comparison of several modern probabilistic logic programming languages like PRISM, ICL and ProbLog. The authors select a probabilistic logic programming language based on the results of this comparison. To implement probabilistic inference in a weakly formalized knowledge base we have selected ProbLog language (ProbLog2 in particular) that is a probability extension of Prolog.

3. Natural language user interface of a dialogue system [№3 за 2016 год]
Authors: Posevkin R.V., Bessmertny I.A.
Visitors: 6386
Many people interact with software every day. Due to computer systems expansion to all fields of people’s activity the problem of transition from visual and command interfaces to natural language user interfaces is becoming more and more relevant. The article describes computational linguistics and natural language processing methods. It considers all phases of natural language processing such as morphological, syntactic and semantic analysis. Limited language is represented as a subset of natural language. This language allows avoiding linguistic uncertainty and decreases natural language elements processing time. Translation system is a part of a dialogue system. Methods of natural language manipulation are applied in machine translation software systems, search engines, data exchange, automatic abstracting and expert systems. The paper represents a formal definition of a dialogue system as a Markov decision process. The authors developed a prototype of natural language user interface for a structured data source. It converts natural language user’s query into SQL query to database. User interface interacts with a database that contains information about existent program libraries and frameworks. Consequently, using natural language processing methods makes it possible to develop natural language user interface providing a capability to interact with a dialogue system.

4. On some problems in the decision support domain [№3 за 2016 год]
Author: Tikhanychev O.V.
Visitors: 6983
The article analyzes the current state of a decision support domain: both in terms of classification principles of such systems and their official definition. This situation complicates the process of creation and application of decision support systems. To normalize this domain the author formalizes a description of a decision-making process and formulates the concepts of “small” and “extended” decision-making cycles. “Small” cycle refers to the process of finding an optimal solution in terms of a set objective function and the restrictions. “Big” cycle includes searching for a non-trivial solution, which cannot be obtained without modifying the objective function or restriction cancellation. Based on this approach the author proposes to clarify the definition of an automated decision-making system and its classification. The main indicator in clarifying the classification is functional defined by the type of a decision-making cycle. Within a new classification it is proposed to replace the currently used class “relative to the user level” by the class “by the functional”, which includes information, calculation and information and intelligent systems. The proposed classification completely corresponds to the modern understanding of automated decision-making support system and can accelerate the process of implementing such systems in management practice.

5. Imprecise temporal information processing [№3 за 2016 год]
Authors: Fomin V.V., Maltsev S.V.
Visitors: 4190
The article presents the results of extensive research resulting from a practical need in improving the effectiveness of representation methods, extraction algorithms, processing fuzzy temporal data. The authors focus on ontological models of prediction and interpretation during interaction with imprecise temporal information. They propose a data storage methodology as well as models and unified procedures for extracting fuzzy information. The pragmatic of temporal data representation algorithms is considered using a new so-called possibilistic temporal information in analytical systems, where the parameters of inaccuracy and efficiency are crucial for making business decisions. The article proposes solutions for the problems of imprecise temporal information storage and retrieval by constructing a model of processing fuzzy temporal relations between time points and/or intervals. The concept of fuzzy temporal information analysis and presentation is described within development of temporal relational databases framework taking into account the criteria of inaccuracies, ambiguity and uncertainty. The authors developed he algebra of possibilistic relations between points and/or intervals. It defines the concept of possibilistic time point, possibilistic interval and corresponding values of possibilistic time points and intervals. The example of extension of one of the relations describes the fragment of a formalized algorithm of algebra implementation. The authors propose an approach to design and modeling a fuzzy temporal domain as well as the relations between possibilistic intervals and time points. The article describes an implemented library of algorithms for processing relations between possibilistic time points and/or intervals, as well as corresponding procedures and functions on SQL platform. At the end, the article describes the results of integration of the library of algorithms in a commercial project of analysis system for a telecommunication company. It also describes areas for further development of application tools based on the theory of a fuzzy-temporal model.

6. Database security: problems and prospects [№3 за 2016 год]
Authors: Poltavtseva M.A., Khabarov A.R.
Visitors: 15187
Data protection issues are critical when ensuring modern enterprise systems security. The paper is devoted to the current situation in database systems security. The paper provides an overview of existing publications, both Russian and foreign authors. It reviews the stages of a database information security systems evolution. The paper describes modern complex problems in DBMS data protection, both theoretical and practical. The analysis of security issues takes into account new directions of the database development, such as “in memory” and NoSQL DBMS. The article describes database features in terms of information security compared with other types of software. It includes the basic database vulnerabilities division based on selected features. The authors formulate promising ways of research and development of security systems for building reliable and secure information management systems.

7. Modeling information security threats in the automated control system for crucial objects on the basis of attack scenarios [№3 за 2016 год]
Authors: Drobotun E.B., Tsvetkov O.V.
Visitors: 9233
Development and construction of information security threat models in various automated and data-processing systems is one of key steps in ensuring information security in these systems and their safe and proper operation. Based on the information security threat model we can define the requirements to information security in automated data-processing systems. We can also define a set of organizational, technical, organizational and technical measures to ensure information security in automated and computer systems for various purposes. The article proposes a method of constructing a threat model for automated control systems of crucial and potentially dangerous objects. The method is based on modeling information security violator’s possible action scenarios. The proposed method takes into account a multilevel structure modern automated control systems for crucial objects and the possibility of implementing the same information security threats using different scenarios by different violators. The proposed methodology would reduce the time for building a model of information security threats in automated control systems for crucial and potentially dangerous objects.

8. Synthesis of protection systems of automated control systems against destroying program influence [№3 за 2016 год]
Author: Drobotun E.B.
Visitors: 5272
Informatization and automation of various industrial facilities (including facilities that create potential safety threats) has made functioning safety of such objects one of the most important issues of modern age. Informatization and automation cause emergence of new types of threats for information security directed to violation of functioning of production objects that create a potential threat of safety, which can be exposed to destructive information and technical influences. These factors in combination with the increasing level of society dependence on functioning stability of the objects, which create potential safety threat, demand a package of measures (including technical measures) to decrease the risk of violation or complete cessation of such objects’ functioning under information and technical influences. The article presents one of possible approaches to constructing rational (in the context of ensuring the demanded security level and minimum influence of protection system on a protected automated system) system of protection against destroying program influences as the main type of possible information and technical influences on automated control systems for crucial and potentially dangerous objects.

9. Search algorithm of optimal sensors location for solving the space monitoring problem [№3 за 2016 год]
Authors: Kochkarov A.A., atskin D.V.
Visitors: 4979
The paper considers the task of space monitoring, an algorithm and software package to solve it. From this problem the authors move to the problem of detection, and then to the problem of geometrical sensors location. It is proposed to use a decentralized robotic network to solve th у proposed problem. The authors set limitations and assumptions for sensors location problem leading to the problem of space covering. They also discretize problem parameters. The paper describes the problem in detail from a mathematical point of view, develops the algorithm to solve it, estimates its complexity. There mathematical and software models of the problem. The authors develop a software package that implements the specified algorithm. The map of the area and a number of physical parameters of the sensor serve as input data for a software package. The package allows calculating points in space, where the placed devices (detectors) help to solve the monitoring problem. This software package was used for a series of tests. The tests showed efficiency, correctness and optimality of the developed approaches and the algorithm. Due to the low polynomial computational complexity of the algorithm the software package can solve monitoring tasks for the particular case of large areas of monitoring, as well as for a big number of sensors.

10. Bee colony optimization algorithm for wireless network planning problem solving [№3 за 2016 год]
Authors: Skakov E.S., Malysh V.N.
Visitors: 6344
The article is devoted to solving a wireless network planning task. The authors defined the problem of base stations’ location, taking into account the level of attenuation of signal during propagation, the level of inter -cell interference and several types of base stations. A problem solution is represented as a vector of structures, each structure stores data about one potential site (an installed base station type, a list of connected clients). A modification of one of metaheuristic methods (bee colony optimization algorithm) has been developed. The bee colony optimization method is a multi-agent method based on simulation of intelligent behavior of agent colonies (the so-called swarm intelligence methods). Naturally, some groups of social insects (for example, colonies of ants, bees, termites) have such intellect. The proposed algorithm is based on the concept of current solution neighborhood. Little changes in current wireless network configuration are used to form the neighborhood. To obtain a new solution from the neighborhood of the current solution we can use one of the following two operations: changing the type of a station to more cheap or removing a base station. The developed algorithm is implemented in Delphi programming environment. The article shows that the new algorithm helps to find the solution for the base station location problem in a reasonable time, much faster than the exact method of exhaustive search. The authors found out how the quality of the problem solution obtained by bee colony optimization algorithm depends on values of control parameters.

| 1 | 2 | 3 | 4 | Next →