IMO.Net Artificial Neural Networks: an object-oriented reusable software component library to integrate Matlab Neural Networks functionality

Author

Méndez, Arturo J. ; Rosello, E.G. ; Lado, María ; Dacosta, Jacinto G. ; Torres, David M. ; Cota, Manuel P.

Author_Institution

Dept. of Comput. Sci., Vigo Univ., Ourense

fYear

2006

fDate

Sept. 2006

Firstpage

159

Lastpage

163

Abstract

We describe a reusable software component library which encapsulates the functionality of the Matlab Neural Networks Toolbox (MNNT). Our library allows the integration of the neural networks toolbox in the software applications development, independently of the platform and development tools used to build it. This library offers two different sets of classes: 1) an API (application program interface) similar to the Matlab toolbox; and 2) a fully object-oriented designed AP. The benefits of using this software component are: 1) to build reusable applications; and 2) to take advantage of the MNNT functionality, power, and robustness

Keywords

application program interfaces; data encapsulation; neural nets; object-oriented programming; software libraries; software reusability; IMO.Net; Matlab neural networks functionality; Matlab neural networks toolbox; application program interface; artificial neural networks; object-oriented reusable software component library; Application software; Artificial neural networks; Computer science; Mathematical model; Neural networks; Object oriented modeling; Robustness; Software libraries; Software reusability; Software tools;

fLanguage

English

Publisher

ieee

Conference_Titel

Computer Science, 2006. ENC '06. Seventh Mexican International Conference on

Conference_Location

San Luis Potosi

ISSN

1550-4069

Print_ISBN

0-7695-2666-7

Type

conf

DOI

10.1109/ENC.2006.18

Filename

4020875