CONSTRUCTION PRINCIPLES OF COMPUTING UNITS FOR INTELLECTUAL DATA PROCESSING
Abstract
An approach to the development of problem-oriented parallel computers for a wide range of tasks of intelligent data processing is described. A high level architecture is shown, where Data Mining algorithms are considered as functional elements and a general or specialized co-processor as an accelerator for computer performance. An accent has been made on the necessity to provide compatibility of the librarian algorithms and the architecture of the parallel co-processor. The technique for correct comparative evaluation of the alternative hardware platforms in the framework of the proposed approach is presented.
About the Author
M. M. Tatur
Белорусский государственный университет информатики и радиоэлектроники
Russian Federation
Russian Federation
References
1. Data Mining: A Knowledge Discovery Approach / K.J. Cios [et al.]. – Springer, 2007.
2. Том, И.Э. Методы интеллектуального анализа многомерных данных для решения задач классификации / И.Э. Том, Н.А. Новоселова, О.В. Красько. – Минск : ОИПИ НАН Беларуси, 2011. – 233 с.
3. Кирсанов, Э.Ю. Нейрокомпьютеры с параллельной архитектурой / Э.Ю. Кирсанов. – М. : Радиотехника, 2004. – 221 с.
4. Аляутдинов, М.А. Нейрокомпьютеры: от программной к аппаратной реализации / М.А. Аляутдинов, А.И. Галушкин, П.А. Казанцев. – М. : Горячая линия – Телеком, 2008. – 152 с.
5. Реализация искусственных нейронных сетей в НТЦ «Модуль» // Компоненты и технологии. – 2005. – № 4. – С. 98–102.
6. Synthesis and Analysis of Classifiers Based on Generalized Model of Identification / D. Adzinets [et al.] //Advances in intelligent and soft computing. – Springer, 2010. – Vol. 71. – P. 529–536.
7. Tatur, M. Problem-Oriented Processors for the Solving of Classification Tasks / M. Tatur // J. of Information, Control and Management Systems. – 2013. – Vol. 11, no. 2. – P. 155–164.8. Cognitive information processing based on a parallel processor / N. Verenik [et al.] // Proc. of 10th Intern. Conf. on Digital Technologies. – Žilina, 2014. – P. 367–371.
8. Wolfram Mathematica [Electronic resource]. – Mode of access : http://www.wolfram.com/ mathematica. – Date of access : 12.01.2015.
9. Weka 3: Data Mining Software in Java [Electronic resource]. – Mode of access : http://www.cs.waikato.ac.nz/ml/weka. – Date of access : 12.01.2015.
Review
For citations:
Tatur M.M. CONSTRUCTION PRINCIPLES OF COMPUTING UNITS FOR INTELLECTUAL DATA PROCESSING. Informatics. 2015;(1):39-44. (In Russ.)
Views: 817
Email this article 