Author(s)

M.C.A. Costa and N.F.F. Ebecken

Abstract

Data mining high performance computing using neural networks M. C. A. Costa and N. F. F. Ebecken COPPE/UFRJ, Brazil Abstract This work discusses the implementation of a neural network data mining strategy in high performance computers. The presentation focuses the study of an application including all the aspects related to the data preparation, the neural network implementation and training. Performance assessment of the implementation is also addressed for two different computer architectures. Some conclusions and recommendations from this experience were done. 1. Introduction In the last few years the development of methods and algorithms for Knowledge Discovery in Databases (KDD) has been growing very fast [1]. This technology involves the process of finding and interpreting patterns from data extracted of huge and complex databases. The Data Mining phase of this technology describes efficient and intelligent algorithms and techniques that handle such data, analyzing

Keywords