Bulevskij faktornyj analiz na osnove attraktornoj nejronnoj seti i nekotoryje ego priloženija

0360031 - ÚI 2012 RIV RU rus J - Článek v odborném periodiku
Frolov, A. A. - Húsek, Dušan - Polyakov, P.Y.
Bulevskij faktornyj analiz na osnove attraktornoj nejronnoj seti i nekotoryje ego priloženija.
[Bulev factorial analysis by means of attractor neural network and its some appendices.]
Nejrokomp'jutery: razrabotka, primenenie. -, č. 1 (2011), s. 25-46. ISSN 1999-8554
Grant CEP: GA ČR GAP202/10/0262; GA ČR GA205/09/1079
Výzkumný záměr: CEZ:AV0Z10300504
Klíčová slova: Boolean factor analysis * recurrent neural network * Hopfield-like neural network * associative memory * unsupervised learning
Kód oboru RIV: IN - Informatika
http://www.radiotec.ru/catalog.php?cat=jr7&art=8576

Obyčnoj zadačej, vstrečajuščejsa pri analize bolšich objemov dannych, javlaetsa poisk ich adekvatnogo predstavlenija v prostranstve menšej razmernosti. Odnim iz naiboleje effektivnych ispolzujemych dla etogo metodov javljaetsa faktornyj analiz. V nastojaščej rabote my predlagaem ispolzovať v kačestve metoda bulevskogo faktornogo analiza attraktornuju nejronnuju seť tipa Chopfilda. Osobennosti funkcionirovania predlagaemoj nejronnoj seti objasnajutsa šag za šagom na primere bulevskogo faktornogo analiza iskusstvenno sozdannogo massiva dannych. V zaključenie my demonstrirujem effektivnosť metoda v priloženii k analizu rezultatov golosovanija v Gosudarstvennoj dume RF i analizu statej, predstavlennych na Meždunarodnoj konferencii po nejronnym seťam.

The usual problem meeting at the analysis of great volumes of data, search of their adequate representation in space of smaller dimension is. One of the most effective methods used for it is the factorial analysis. In the present work we suppose to use as a method Bulean the factorial analysis attractor of the Hopfield neural network. Features of functioning of an offered neural network speak step by step an example Bulean the factorial analysis is artificial the created data file. Efficiency of a method in the appendix to the analysis of results of voting in the State Dumas of the Russian Federation and is shown to the analysis of clauses presented on the International conferences on neural networks.
Trvalý link: http://hdl.handle.net/11104/0197681