   |
  |
||
|
|||
 |
|||
“Systems and Means of Informatics” scientific journalVolume 26, Issue 4, 2016Content Abstract and Keywords About Authors METHOD OF INCREASING INFORMATION PERTINENCE FOR E-COMMERCE RECOMMENDER SYSTEMS BASED ON IMPLICIT DATA.
Abstract: The paper describes the method of increasing pertinence of information in e-commerce recommender systems based on implicit data, i.e., due to the processing of user activity associated with the decision-making process. The method works successfully in situations where information about user activity is absent or little informative. Practical application of this method in e-commerce systems can improve their efficiency through targeted supply of goods and services to consumers. The main feature of the proposed method is the combined use of Item-Item CF (collaborative filtering) and User-User CF methods taking into account the implicit data collected. The features of the proposed method are verified by a prototype software that is installed on the existing online store Thaisoap. Keywords: pertinence search; collaborative filtering; e-commerce recommender system; implicit data targeting
MODELS OF FAULT-TOLERANT SELF-TIMED CIRCUITS .
Abstract: The development of both the reliable and the energy-efficient computing systems is one of most important tasks in the XXI century. The usage of self-timed circuits makes it possible to improve energy-efficiency of a computing system. However, the complexity increase in not fault-tolerant self-timed circuits leads to decrease of reliability. The fault-tolerant self-timed implementation of digital devices makes it possible to increase not only reliability but also the complex index "energy-consumption/reliability." The further development of synthesis methods of fault-tolerant self-timed circuits will allow to compensate the negative effects of self-timed circuit's development by the positive effect of complex index increasing. The paper describes the models and algorithms of fault detection which are developed to improve validity and reliability of actively fault-tolerant self-timed circuits. Keywords: reliability; energy-efficiency; energy-reliable; self-timed circuit; fault-tolerant; self-repair
SECURITY EVALUATION IN SECURE ARCHITECTURE OF DISTRIBUTED INFORMATION SYSTEMS.
Abstract: The necessary condition of secure architecture of distributed information systems is the possibility of effective enumerating of information flows and the analysis of their contents. Completeness of the analysis of security of distributed information systems comes down to completeness of enumerating of information flows and possibility of the analysis of their contents. Contradictions of secure architecture of distributed information systems can arise if there is a forbidden information flow, or a ban on an allowed information flow, or an admission of failure is possible. The paper gives an example of architecture of distributed information systems in which the analysis of interactions of components represents a complex computing problem and an example of secure architecture when it is possible to effectively analyze all interactions in distributed information systems. Keywords: secure architecture of distributed information systems; methods of the analysis of security of distributed information systems
ON SYSTEM-TECHNICAL AND REGULATORY-METHODOLOGICAL PROBLEMS OF DATA SECURITY IN CLOUD AUTOMATED INFORMATION SYSTEMS USING ARTIFICIAL INTELLIGENCE TECHNOLOGIES.
Abstract: The article focuses on information security in cloud automated information systems (AIS) and the possibility of using artificial intelligence in data protection systems. Major problems of data protection in cloud computer systems are analyzed. General components of a data protection system are considered. Presently, artificial intelligence is used in data protection systems to some extent. The authors propose to extend the area of application of artificial intelligence in order to protect data. The article contains an analysis of the current regulatory base regarding information security in the cloud computer systems. Keywords: automated systems; information security; cloud computer systems; artificial intelligence; standards (norms); data security threats
METHODOLOGY OF CREATING WEB-SERVICE INTERACTIONS IN THE SYSTEM OF DISTRIBUTED SITUATIONAL CENTERS.
Abstract: The article considers the questions of information interaction between components of a system of distributed situational centers. The authors propose an architecture of the system's interaction based on SOA (service-oriented architecture) using the web technology as the underlying instrument for service invocation, as well as the expansion and development of business processes through the application of the EDA (event-driven architecture) technology and the organization of interaction of the system's components using the ESB (enterprise service bus). Keywords: center; service; SOA; EDA; ESB; web service
APPLICATION OF THE CUDA ARCHITECTURE FOR IMPLEMENTATION OF GRID-BASED ALGORITHMS FOR THE METHOD OF MOVING SEPARATION OF MIXTURES.
Abstract: The paper presents the implementation of the grid methods for finding maximum likelihood estimators in the mixed probability models based on the software solutions for the computations on GPUs using the NVIDIA CUDA technology. The hierarchy of programming classes is described, an approach to the initial estimation and further modification of the parametric grids is proposed, and the convergence speed and other characteristics of the developed methods are examined by the test data sets. The key characteristics of the method including the change of approximation error by the l1 metric and reducing the number of components under the iterative steps are demonstrated by the graphs.
Keywords: CUDA; GPU; grid methods; mixed probability models; moving separation of mixtures; online software
PROFIT MAXIMIZATION IN G/M/1 QUEUING SYSTEM ON A SET OF THRESHOLD STRATEGIES WITH TWO SWITCH POINTS.
Abstract: The problem of maximizing the average profit per time in G/M/1 queuing system is considered on a set of stationary access restriction threshold strategies with one "switch point." The objective function depends on the following measures: service fee, hardware maintenance fee, cost of service delay, fine for unhandled requests, and fine for system idle. The authors have formulated the necessary conditions of existence of finite problem solution on a subset of threshold strategies with fixed distance between the upper and lower thresholds and have got necessary and sufficient conditions for optimality of threshold strategy on this subset. The authors have also developed a method of finding the optimal strategy and algorithm for calculating the parameters of the optimal strategy and the corresponding value of the objective function. Keywords: queuing system; threshold strategy; optimization
METHOD OF WEIGHTED DISCRIMINANT SYSTEMS FOR THE CLASSIFICATION OF OBJECTS WITH MISSING DATA.
Abstract: The authors describe a method of classification of objects with missing or partially known data. The tasks related to the processing of incomplete data are common in medicine. Patient data can contain gaps or missing. Information classification of patients with varying degrees of schizophrenia was carried out using the new method. Schizophrenia is a genetic disease; so, important is the task of studying the genetic predisposition of a person to the disease. Analysis of associations between polymorphisms of genes was performed. A distinctive feature of the provision of medical data is its emptiness by more than 70%. The dimension of discriminant signs was significantly reduced and high reliability of forecasting has been received. Keywords: discriminant analysis; censored data; classification functions; genetic diseases
VISUAL REASONING MODELING IN A FUNCTIONAL HYBRID INTELLIGENT SYSTEM.
Abstract: Hybrid intelligent systems have been successfully used to solve the problems with high modeling complexity due to combination of multiple methods of simulation of human intellectual activity and compensation of some methods' shortcomings by advantages of the other. Most hybrid intelligent systems are based on the interaction of elements of the symbolic-logical model of knowledge, which limits their opportunities in comparison with the groups of experts, operating character-logical and visual-imaginative knowledge. Modeling of the collective visual-imaginative reasoning is proposed for implementation in a new class of intelligent systems - hybrid intelligent systems with heterogeneous visual field. To implement such reasoning, the formalized model of visual languages, based on the concept of the semiotic system by D. A. Pospelov, and the principles of their combination in the heterogeneous visual field are proposed. Keywords: hybrid intelligent system; heterogeneous visual field; visual language; semiotic system
ALGORITHMIC SOFTWARE OF THE OPTIMIZATION PROBLEM FOR CARGO DELIVERING BY WATER TRANSPORT.
Abstract: The paper considers the problem of constructing an optimal route of homogeneous cargo delivery from a family of producers to a family of consumers with the location of points on two riverbanks by a ship with limited capacity. Problems of this type arise, in particular, for the delivery of food and essential items to flood victims. The corresponding mathematical model is constructed. Several algorithms are suggested for solution of this problem (the exact algorithm of branch and bound, the greedy algorithm and two original heuristic algorithms that take into account the specific features of the problem). A computer experiment on simulated data was conducted using the developed software.
Keywords: transportation logistics; optimization; heuristic methods; branch and bound method; greedy algorithm
CROSS-LINGUAL DATABASE FOR ANNOTATING LOGICAL-SEMANTIC RELATIONS IN THE TEXT.
Abstract: The problem of designing a cross-lingual database is described. The purpose of such a database is annotating logical-semantic relations between fragments of parallel texts in two or more languages. One of the objectives of its design is the information and computer provision of constructing by linguists a classification scheme of logical and semantic relationships not depending on the text language. A point of this design problem is that linguists annotate logical-semantic relations using a list of rubrics, which is being formed in the process of annotating by means of the cross-lingual database. According to its functions, the database can by classified as a supracorpora database. Its pilot version allowed linguists to form, at the same time, thousands of annotations and a list of rubrics to be used in the annotating process. Keywords: cross-lingual databases; annotating; parallel texts; corpus linguistics; logical-semantic relations
SOCIAL POTENTIAL OF INFORMATION TECHNOLOGIES IN MODERN RUSSIA.
Abstract: The article analyzes the social potential of information technologies in the emerging global information society. The article describes the structure of this potential and the role of its individual components in the solution of actual problems of social and economic development of Russia. The article shows the possibility of a more efficient use of this potential in the new Strategy of scientific and technological development of the Russian Federation in the long term. Keywords: information society; information culture; information technology; quality of life; economy and work; education and health care; public safety; international communication
CREATION OF A HUMAN-MACHINE ENVIRONMENT FOR PROBLEM SOLVING.
Abstract: The article concerns the issue of creating a human-machine environment for solving the problems of modeling in science and engineering, computer-aided design, control, distance learning, etc. The authors' results related to the problem are presented. Keywords: human-machine environment for solving problems; programmable machine; task constructive object; mandatory and orienting rules; online service
CONCEPTUAL BASIS FOR THE UNSTRUCTURED INFORMATION AUTOMATED PROCESSING IN PERSPECTIVE CONTROL SYSTEMS.
Abstract: The knowledge-based approach to unstructured information automated processing is considered. This approach is positioned as theoretical basis for the transition from data-processing industry to the industry of knowledge extracted from unstructured texts in natural language processing. The methods for constructing a linguistic and ontological processor and semantic search engine in perspective automated control systems (ACS) have been proposed. Keywords: unstructured information; knowledge-oriented approach; NL-text; linguistic and ontological processor; semantic search engine
|
 | |
 Phone of the Center: +7 (499) 135-62-60E-mail of the Center: ipiran@ipiran.ru | 
RUS
|