Национальный цифровой ресурс Руконт - межотраслевая электронная библиотека (ЭБС) на базе технологии Контекстум (всего произведений: 634617)
Контекстум
.
0   0
Первый авторBolotova
АвторыMakhortov S.D.
Страниц4
ID404421
АннотацияThe relevant LP-inference, which is based on the solution of logical equations, is an effective tool that can be used for research and optimization of production-logical systems. It significantly reduces the number of executed queries to an external source of information (either to a database or an interactive user). The preference is given to testing the facts that are really needed in the inference. However, experiments have shown that the process of using the relevant LP-inference may require an excessive amount of computational resources of the computer. So, the relevant LP-inference method was modified to use parallel computing algorithms. This paper describes the implementation of a multi-threaded algorithms for relevant LP-inference and provides the pseudocodes of these algorithms. Multi-threading is a fundamentally new element in the implementation, which allows speeding up the process of constructing sets of facts that are required in the inference, and their further processing.
УДК519.688
Bolotova, S.Yu. The Algorithms of the Mulri-Threaded Relevant LP-inference / S.Yu. Bolotova, S.D. Makhortov // Вестник Российского университета дружбы народов. Серия: Математика, информатика, физика .— 2014 .— №2 .— С. 218-221 .— URL: https://rucont.ru/efd/404421 (дата обращения: 20.04.2024)

Предпросмотр (выдержки из произведения)

Bolotova, S. D. Makhortov Voronezh State University 1, Universitetskaya pl., Voronezh, Russia, 394006 The relevant LP-inference, which is based on the solution of logical equations, is an effective tool that can be used for research and optimization of production-logical systems. <...> It significantly reduces the number of executed queries to an external source of information (either to a database or an interactive user). <...> The preference is given to testing the facts that are really needed in the inference. <...> However, experiments have shown that the process of using the relevant LP-inference may require an excessive amount of computational resources of the computer. <...> So, the relevant LP-inference method was modified to use parallel computing algorithms. <...> Multi-threading is a fundamentally new element in the implementation, which allows speeding up the process of constructing sets of facts that are required in the inference, and their further processing. <...> Key words and phrases: backward inference, relevant inference, logic equations, parallelism, multi-threading. 1. <...> Introduction in the field of artificial intelligence. <...> They are used in the theory of learning and problem solving systems, the development expert systems, covering a large variety of applications in several areas, such as medicine, engineering, exploration, and others [1]. <...> Production systems have very rich history and the fact that they are still of interest is confirmed by numerous modern applied and theoretical researches in this area [2–5]. <...> However, production system programs are very computationally intensive and run Production systems is an important area of the theoretical and applied research quite slowly. <...> In particular, they often require an exponential number of computation and intensive exchange with external memory. <...> This fact reduces their effectiveness, and therefore, many experts try to avoid working with programs that require extreme amounts of resources [1]. <...> Consequently, a significant increase of production systems performance is the important part in theoretical and applied research. <...> There are several possible ways for speeding up the execution of production systems. <...> Multi-threading is a fundamentally new element in the implementation. <...> A General Description of the Algorithm based on the solution of the production logical equations and for this purpose it uses mathematically-based algorithms for finding <...>