Harvester open publications of NAS Ukraine
Harvester open publications of NAS UkraineNot a member yet
2350 research outputs found
Sort by
Experience of the semantic technologies use for intelligent Web encyclopedia creation (on example of the Great Ukrainian Encyclopedia portal)
No full textWe consider the conceptual principles of the Great Ukrainian Encyclopedia development and analyze the main advantages of its portal version – e-VUE. This portal version requires the relevant technological platform with high-level expressiveness and efficiency. Development of e-VUE with complex knowledge structure causes the study of modern Web-technologies and matching their capabilities with the specific requirements of encyclopedia. Therefore we consider in details the project idea and purposes, the specifics of e-VUE information representation (number and volume of articles, content types and sources, relations between content elements and their properties, business processes of article publication, design requirements) and processing (navigation, requests, integration and matching means) to estimate the Wiki-technology feasibility use (namely - opportunities and restrictions of MediaWiki technological platform). Also attention is paid to the means of information security ensuring and semantic-based establishment of content access rights. The results of such analysis show that MediaWiki with semantic plugins can be used as a base for e-VUE portal construction but needs in extension of its functionality based on modern knowledge management tools and formats of their interoperable representation. Now we are oriented on use of ontological analysis and Semantic Web standards (OWL, RDF, SPARQL). In terms of the Semantic Web, e-VUE is a distributed database with heterogeneous types of information objects that can be used both by humans and by external software (e.g. by intelligent agents). The main principles of e-VUE knowledge base organization are grounded on Wiki categories and semantic properties of Semantic MediaWiki that are used for formal representation of the e-VUE typical information objects (TIO). Such approach allows to express TIO relations, properties and characteristics that can be used by logical inference and semantic retrieval to improve the navigation between portal elements and integration of data from various Wiki pages.Problems in programming 2020; 2-3: 246-25
Domain engineering approach of software requirements analysis
Get PDFRequirement analysis is one of the important processes in software development lifecycle management. In Agile approach requirements software models are the basic of generating other software development artifacts. Improving requirements approaches and techniques allows avoiding mistakes in other software development artifacts. Domain engineering fundamentals is the basic for “template oriented” approaches of software development artifacts designing. Reusing domain models and knowledge allows adding details in vertical “model to model” transformation operations, refine generated software development artifacts, organize systematic software reuse and perform many other activities. Paper proposes an approach of requirement analysis based on UML Use Case diagrams transformations into communication ones and the next refinements of them by means of information from domain models. The advantages of the proposed approach is the next: proposed transformation method involves ”many to many” transformation in order to save the semantic of initial model. Domain knowledge are used to complete communication diagram by means of adding details after transformation to them. In order to perform Use case to communication transformation graph representation of software models is chosen.Problems in programming 2020; 2-3: 164-17
Construction of diagnostic expert-medical system using neural networks
No full textIn the article was discussed the methods (decision trees, deep learning algorithms, k-nearest neighbors, neural networks) to create diagnostic expert medical systems. For practice part were developed diagnostic API based on chosen classifiers that implement the algorithms and a study of their work was conducted. Namely, classifiers based on neural networks, decision trees and k-nearest neighbors method were compared. The parameters for the selected classifier were optimized. As a result, were selected parameters on which the data were researched. In addition, the dataset of information of patients who had heart attack was researched to develop a diagnostic system for revealing heart diseases. The diagnostic API for revealing patients’ heart diseases is described. Keywords: diagnostic systems, medical systems, neural networks, decision trees, diagnostic API.Problems in programming 2020; 2-3: 384-39
The description of lists and sets of meta-language of normal forms of knowledge
No full textLists use for representation of various knowledge. As lists it comfortably to present formulas, functions, trees, columns, great numbers and many other difficult objects. Great number - one of the most essential structures of data, used both in mathematics, and in programming. The formalization of lists, list-based predicates and set-based predicates in the meta-language of normal forms of knowledge is presented, based on the known Prolog-formalizations of these concepts, which use a list-domain. Among the described list-based predicates are the following: adding an element to the list, removing an element, finding the last element of a list, finding adjacent elements in a list, concatenation of lists, reversing a list, palindrome, etc. Using the list-domain, the set-based predicates are described as follows: converting a list into a set, checking if an element is in a set, concatenation, intersection, difference, symmetrical difference, identity, complement of sets, relation of subset, proper subset. Problems in programming 2020; 1: 03-1
Parallel algorithms optimization using Actor Model
No full textIntroduced methods and instrumentation tools for actor model applied to block recursive algorithms optimization. Created formal model of distribution and coordination of the tasks in computation cluster as asynchronous reactive processes with message-passing represented with an actor model and choreography of actors. Created declarative definitions of algorithms which compiles to the system of actors. Proposed scheme of data placement in a cluster using prioritization of block-recursive operations to reduce idling time, data movement, with increased parallelism in situation of high-speed processors and reduced network bandwidth. Implemented adaptive adjustment of the data placement in a cluster at run time to account for current cluster load. Created autotuning of the actor placement in а cluster which uses statistics of previous runs for optimization. Usage of choreography of actors allows to remove central coordinating element and to avoid hard dependencies between cluster nodes, which provides flexible data placement, improves fault tolerance with no single point of failure and allows to use self-healing. Problems in programming 2020; 2-3: 126-137
Causal inference from data. On some inadequacy problems of structures with hidden causes
No full textThe reliability of causal inference from data (by independence-based methods) is analyzed. We uncover some mechanisms which may result in model inadequacy due to sample bias and hidden variables. We detect some specific problems in recognition of direction of influence when some causes are hidden. Incorrectness of known rule for edge orientation (under causal insufficiency) is revealed. We suggest the correction to the rule aiming to retain model adequacy.Problems in programming 2020; 2-3: 392-40
Elements of concrete algorithmics: computability and solvability
No full textAn approach to proving the fundamental results of the theory of recursive functions using specific algorithms is consider. For this, the basic constructions of the algorithm are describing exactly and Church's thesis for more narrow classes of algorithmically computational functions is specified (concretized). Using this approach, the belonging of functions to classes of algorithmically computable is argued by the construction of the corresponding algorithms.Problems in programming 2020; 2-3: 198-20
Automated design and parallelization of programs for heterogeneous platforms using algebra-algorithmic tools
No full textMethods and software tools for automated design and generation of OpenCL programs based on the algebra of algorithms are proposed. OpenCL is a framework for developing parallel software that executes across heterogeneous platforms consisting of general-purpose processors and/or hardware accelerators. The proposed approach consists in using high-level algebra-algorithmic specifications of programs represented in natural linguistic form and rewriting rules. The developed software tools provide the automated design of algorithm schemes based on a superposition of Glushkov algebra constructs that are considered as reusable components. The tools automatically generate code in a target programming language on the basis of the specifications. In most computing problems, a large part of hardware resources is utilized by computations inside loops, therefore the use of automatic parallelization of cyclic operators is most efficient for them. However, the existing automatic code parallelizing tools, such as Par4All, don’t account the limited amount of accelerator’s onboard memory space while real-life problems demand huge amounts of data to be processed. Thus, there is a need for the development of a parallelization technique embracing the cases of massive computational tasks involving big data. In the paper, a method and a software tool for semi-automatic parallelization of cyclic operators based on loop tiling and data serialization are developed. The parallelization technique uses rewriting rules system to transform programs. The framework for parallelization of loops for optimization of computations using graphics processing units allows semi-automatic parallelization of sequential programs. The approach is illustrated on an example of developing a parallel OpenCL image convolution program.Problems in programming 2020; 2-3: 103-11
Neural networks’ learning process acceleration
Get PDFThis study is devoted to evaluating the process of training of a parallel system in the form of an artificial neural network, which is built using a genetic algorithm. The methods that allow to achieve this goal are computer simulation of a neural network on multi-core CPUs and a genetic algorithm for finding the weights of an artificial neural network. The performance of sequential and parallel training processes of artificial neural network is compared.Problems in programming 2020; 2-3: 313-32
Роль зелених насаджень міста у збереженні диких бджіл (Hymenoptera, Apoidea)
No full textThe aim of our study was to assess the attraction and value of flowering plants at green areas in support and sustenance of wild bees (Hymenoptera, Apoidea) populations in Kyiv. Study objects were the most common flowering ornamental plants of the city and the wild bees visiting their inflorescence during the vegetation season to collect pollen and feed on nectar. The study was conducted at 16 areas of observation and material collection, which include urban parks, M.M. Gryshko National Botanical Garden, green spaces of residential areas, roadsides, etc. Insects were collected following the standard method of catching individual specimens during the spring and summer periods of 2012–2018. Based on the observations, we visualized trophic relations of bees with plants and calculated the biodiversity index of visiting insects for plants. The blossom periods were analyzed using phenological data. Examination of urban green areas revealed ornamental plants that were the most attractive for bees, including more than 35 taxa of 20 families of trees, shrubs, and grassy plants. Bees are superiorly attracted to plants of the genera Rudbeckia, Sedum, Gypsophila, Cerasus, Tagetes, Spiraea, Lonicera, and Aesculus. There is a succession of plant flowering during spring-summer season, which must be considered while planting of greenery. Certain plant species attract insects at each blossom period, for example, Prunus, Rhododendron, Crataegus, Aesculus in spring, most of Asteraceae – in summer. The diversity of blossoming plants is significantly lower at the end of summer and beginning of fall, coinciding with the decreasing flight activity of wild bees. Overall, the studied ornamental plants attract not only the most common species of wild bees but also highly specific and rare species such as Bombus argillaceus and Xylocopa valga, protected by the Red Data Book of Ukraine. We found that blossoming green areas made up of trees, shrubs, and herbs are essential for feeding many species of wild bees and sustaining their populations in urban conditions