Harvester open publications of NAS Ukraine
Harvester open publications of NAS UkraineNot a member yet
2350 research outputs found
Sort by
Optimization of auto-tuning of programs using neural networks
No full textAuto-tuning of programs is a method of self-tuning of internal parameters of the program, affecting its speed, in order to achieve high performance indicators, but it can take a lot of time for testing. In this paper, we propose to improve the method of auto-tuning of programs using neural network algorithms and statistical simulation. The automatic learning of the program model on the results of the "traditional" tuning cycles with the subsequent replacement of some auto-tuner calls with an evaluation from the approximation model allows to significantly accelerate the search for the optimal program variant. Problems in programming 2017; 2: 40-4
Binary Relational Data Model
No full textThe paper is related to the problem of data integration in the Semantic Web and is a continuation of the previously published works, which was dedicated to the creation of mappings from the description logic (DL) into binary relational data model. In this paper, we define the binary relational data model, namely, its structure and algebra. This algebra contains classical relational algebra operations that do not increase arity of relations, modified classical relational algebra operations that increase arity, and a number of additional operations that allow interpreting certain concepts and roles of DL. The publication also discusses how to convert the n-ary relational data structure into a binary one. In this paper, the previously obtained research results are used, namely RM2 data structure and mapping of basic concepts, roles, axioms of DL ALC and its extensions into the relational data model. Problems in programming 2017; 2: 96-10
Development of Meteorological forecasting web portal on multiprocessor platform
No full textThe article deals with the further development of a service-oriented portal solution for providing meteorological forecasting services. In particular, the three-dimensional algorithm for solving the task of circulation atmosphere modeling was implemented using graphics processing units (GPU) as a computing device for forecast calculation as well as for its visualization.Problems in programming 2016; 2-3: 246-25
DS-theory. Scientific aspects and prospects for development
No full textThe work presents the theory of decomposition schemes as a scientific theory. It’s attributes – paradigm, field of research, aim of research, main task, decision method, basic theory model – decomposition scheme were described. Practical aim of the theory is to suggest the mechanism of applied algorithms generation (not machine code). It is shown that the decomposition scheme as a description changes the algorithm description and still remains declarative on the contrary to the algorithm description as imperative. Thereby the decomposition scheme is the description, original for algorithms’ generation. The kinds of decomposition schemes and operations on them have been described. Algorithmically relevant factors which should be considered when generating algorithms in order the algorithms turned real have been described. Working over formalization and mathematical description of the phenomena and objects of the theory of the decomposition schemes have been described here. The mechanism for control of conclusions and results of the theory has been suggested. This work also describes the direction of development of the theory of the decomposition schemes. The priority direction relates to the creation and development of the system of capacious notion and abstractions. From the point of view of the practical application of the theory of the decomposition schemes there the approach which is alike to the one used in machine-building – development and promoting of hi-tech when required – as a result of the targeted research has been suggested. Problems in programming 2017; 1: 03-2
The technique of using Description Logics in the process of constructing a composite service at the functional level
No full textAutomated composition is one of the most difficult web-services tasks. It has two purposes: first is to satisfy complex client’s requirements, second is to reduce a complexity of developing web-service to satisfy complex application system. Created theoretical apparatus is the basis for developing the algorithms of automated solving a semantical web-services composition task. Proposed approaches are based on using description logics. It is effective and powerful tool due to its mechanisms of judgments and the possibilities of logical inference and giving semantic meaning to descriptions.Problems in programming 2018; 1: 77-9
Analysis of formal models and standards for structured electronic document in corporate informational system
No full textPermanent acceleration of development of information technologies and the resulting changes in typical business processes of organizations create new forms of complexity of electronic document flow while creating modern corporate information systems. The main task in mapping structured documents to electronic form is to define the universal mechanism of document markup and to determine its specific information fragments for correct interpretation in software applications. A comparative overview of the existing languages for describing structured electronic documents and the standards on which they are based is provided. The advantages and disadvantages of their use for addressing typical tasks of electronic document flow organization that arise while developing corporate information systems are determined.Problems in programming 2018; 1: 128-14
Transition systems as method of designing applications in GPGPU technology
No full textThe method of researching systems with high-performance computing support, based on the transition systems apparatus (discrete computational model), is proposed. Two variants of synchronous product limitations of transition systems that model tha Nvidia CUDA approach are proposed. transition systems that represent two types of instructions, process of the warp instruction execution, and the process of warp scheduling were described. GPGPU application execution model was formalized and its correctness was proved. Two variants of the relevant Petri net which allowed automatic or semi-automatic detection of design errors were obtained.Problems in programming 2018; 2-3: 012-02
On the possibility of a systematic study of the production management systems
No full textDegree of similarity between systems, subsystems and tasks of MIS for different industries (the companies, departments, shops) is minimum that generates big expenses in designing of the software. The methodology of designing of MIS hides significant reserves for essential reduction in cost of these works. The purpose of this work is in offering the generalized (unified) scheme of tasks or subsystems of MIS that will allow to reduce costs for their designing. The conceptual model of the production management system is offered as such scheme. Basic managerial system with a cluster of models of the simplest production systems, managerial system tree, planning feature, function of cost accounting and the idea of polyfunctionality of production information is attributes of the conceptual model. The method of representation of multi-level a production management system as superposition of two-level managerial systems is offered. Two ideas – cluster of models of the elementary production systems and superposition of managerial systems – create prerequisites for classification and typification of the production management systems. The organizational mechanism and the mechanism of aggregating of production information are objects of a first-priority research among attributes of model of the elementary production systems. Classification and typification of the production management systems are a necessary condition for creation universal (or quasiuniversal) the conceptual model of the production management system. The conceptual model of the production management system is considered as a basis for MIS designing.Problems in programming 2017; 2: 48-6
Semantic Wiki-resources and their use for the construction of personalized ontologies
No full textThe necessity of personalized knowledge use in modern distributed applications and sources of such knowledge are considered. The features Wiki-resources and their semantic markup means are analyzed. Method of semantic Wiki-resource use for generating of personalized ontologies of user tasks is proposed. The means of application of these ontologies in intelligent Web-based applications, in particular for semantic search, are considered.Problems in programming 2016; 2-3: 188-19
Analysis of Genetic Algorithms for solving the 2D Orthogonal Strip Packing Problem
No full textA class of genetic algorithms for solving the 2D Strip Packing Problem is investigated. The theoretical analysis of the complexity of implementing decoders MERA and BLF is done. Original implementations of these MERA and BLF decoders enhanced with a number of heuristic optimizations are proposed. Genetic algorithm for solving the 2D Strip Packing Problem for special cases (allowed/forbidden objects rotation by 90°) with the use of MERA/BLF decoders is proposed. Extensive computational experiments with well-known instances are performed to analyze different configurations of basic parameters of proposed genetic algorithm. The comparison of the obtained algorithm with other known algorithms is given.Problems in programming 2016; 4: 104-11