Title :
Object oriented design methodologies for the implementation of artificial neural networks
Author :
Whittington, G. ; Spracklen, C.T.
Author_Institution :
Dept. of Eng., Aberdeen Univ., UK
Abstract :
Examines the effectiveness of applying object-oriented programming (OOP) methodologies to the synthesis and implementation of artificial neural networks (ANN). Included in the discussion are the range of potential mappings between OOP and ANN methodologies, OOP based development environments for ANN, and OOP based ANN description languages. The effectiveness of each of these strategies and their effect upon the development cycle is explored. The run-time performance penalties incurred by using object-oriented languages and systems is reported and includes a summary of experimental results
Keywords :
neural nets; object-oriented programming; artificial neural networks; description languages; design methodologies; development cycle; object-oriented programming; potential mappings; run-time performance penalties; synthesis;
Conference_Titel :
Neural Networks: Design Techniques and Tools, IEE Colloquium on
Conference_Location :
London
