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

Journal influence

Higher Attestation Commission (VAK) - К1 quartile
Russian Science Citation Index (RSCI)

Bookmark

Next issue

2
Publication date:
16 June 2024

Articles of journal № 2 at 2015 year.

Order result by:
Public date | Title | Authors

21. Ambiguous semantics and incorrectness when working with C# threads [№2 за 2015 год]
Author: Billig V.A.
Visitors: 9016
Modern processors are multi-core. Increasing computing power by increasing the number of pro-cessors and the number of cores in each processor will only be progressing. Changing of hard leads to changing soft. As a result, parallel computing becomes one of the main directions of modern programming. When programming in C#, parallel computing is supported by the mechanism of threads created by an oper-ating system. You can create a thread object of the Thread class and associate a certain piece of code with it. It seems natural that when you create a program object of Thread class, the operating system creates a physical thread which will execute the code. For example, it occurs with file objects – creating a file object in the pro-gram leads to the creation of a physical file. The article shows the situations when the “natural” semantics takes place, as well as the situations when crea-tion of program thread does not create the physical thread. The author provides recommendati ons for optimizing the degree of parallelization in recursive methods and in methods without recursion. The research has made it possible to detect the situation when interaction of two important mechanisms (threads and anonymous methods) leads to incorrect work. The constructed examples show incorrect work of anonymous methods when paralleling.

22. Implementation of a case-based module for intelligent systems [№2 за 2015 год]
Authors: Zo Lin Khaing, Ar Kar Myo, Varshavskiy P.R., Alekhin R.V.
Visitors: 14151
The paper considers important problems of applying a case-based approach in modern intellectual (expert) sys-tems. In particular, in decision support systems focused on assistance to a decision maker when managing complex objects and processes under uncertainty in input data and expert knowledge. Using the mechanisms of plausible reasoning and Case-Based Reasoning (CBR) is directed to expanding the scope of application of intelligent systems under uncertainty as well as reducing time to find solutions. To implement a case-based approach, the authors propose a network (ontological) representation model of cases and hy-brid case retrieving algorithm which is based on the structural mapping theory and the nearest neighbor method. In addition, the paper discusses the possibility of optimizing (reducing) a knowledge base (use case base) of a case-based system (CBR system) using clustering methods. The article proposes CBR system architecture based on the proposed case-representation model and retrieving and opti-mization algorithms for a precedent base (accumulated experience of the system). The authors describe the features of soft-ware implementation of the basic modules of CBR-system prototype in MS Visual Studio 2010 using the Protégé ontology editor for MS Windows. Effectiveness evaluation of the proposed approach and developed software has been carried out on several test data sets offered by the Department of Computer Science, University of California (UCI Machine Learning Re-pository).

23. A problem-oriented editor for design of rule-based knowledge bases [№2 за 2015 год]
Authors: Berman A.F., Grishchenko M.A., Nikolaychuk O.A., Yurin A.Yu.
Visitors: 9793
The paper describes a modification of the universal editor for design of rule-based knowledge bases. The de-scribed modification of the editor is intended for assessment of technical conditions and operating life of petrochemical equipments. The article also presents architecture, main functions and a structure of files for configuration of the editor. The feature of the modified editor is a semantic layer. This layer provides setup of the editor according to the features of subject domain as well as its integration with inference engines. This provides the ability to test developed knowledge bases . The semantic layer is implemented as a set of templates describing facts and rules (cause-and-effect relations) in prob-lems of technical conditions and operation life assessment. The templates allow abstracting from the syntax of knowledge representation languages (programming languages for knowledge bases, in particular, CLIPS – C Language Production Sys-tem). The second part of the semantic layer is the algorithms that provide dynamic creation (generation) of elements of user interface. The editor is used for developing knowledge bases to examine industrial safety of petrochemical equipments. Application of the editor reduces the participation time of a knowledge engineer in the process of design a knowledge base and provides opportunities for experts to complement the knowledge base in the operation of the expert system.

24. Web-oriented component of an expert system [№2 за 2015 год]
Authors: Korshunov S.A., Pavlov A.I., Nikolaychuk O.A.
Visitors: 12468
The paper discusses problems of expert system component implementation in the form of a web application. This component will allow the experts to solve problems using formed logical rules describing the dynamics of studied pro-cesses. Web-based access to the component will enable public access which is especially important in interdisciplinary stud-ies that require a team work of specialists from different areas of expertise. The component has the following architecture: web-service; module interacting with the database and implementing the programming interface to access the database; knowledge management module for working with the rules and facts; rules editor based on RVML-notation; module of communication with web-service that implements the user interface access to the expert system web-service; database for knowledge base storage (facts, rules and initial conditions). The implementation of the web-service is considered in detail. The paper mentions a description of its functions, architecture and functional modules. The main task of a web-service is to prepare database information (facts and rules) for further processing by engine, i.e. convert it from the generalized form to JESS format. The obtained information will be used to implement inference based on rules and facts. The main modules of web-service are: database interface that implements a software interface to access the database; an inference engine that i m-plements the process of reasoning according to the rules and is presented in the form of a software library imp lementing the JESS engine; conversion unit of the rules and facts from the generalized to engine format, the implementation depends on the engine.

25. Simulation of system indicators to estimate the effectiveness of scientific schools [№2 за 2015 год]
Authors: Ilyasov B.G., Karamzina A.G., Fazletdinova Yu.R.
Visitors: 11757
The scientific activity effectiveness of a research university depends on the quality of its function-ing and development control. Such control should be based on knowledge of the current stat e of a research uni-versity. It can be determined by evaluating scientific activity. Practical implementation of the scientific activity evaluation is difficult because of the subjective formation of scientific knowledge and the need to use different methods and indicators engaging experts. To assess scientific activity of a research university the paper proposes to use an integrated assessment of scientific schools effectiveness. A hierarchic system that includes level system of estimating gradients, parameters, system indicators and general integrated assessment is used to calculate i n-tegral evaluation of effectiveness. The authors prove the necessity of creating an automated information system of scientific schools effective-ness evaluation. Its application will allow improving the quality of control at all levels of a research university and providing these levels of control with necessary information. A complex of models using a structural a p-proach to the design of automated information systems is developed. This complex includes dataflow diagrams and diagram “entity-relationship”. A software functional diagram is developed to reflect an interaction of the main subsystems of the automated system.

← Preview | 1 | 2 | 3