Home // ADVCOMP 2014, The Eighth International Conference on Advanced Engineering Computing and Applications in Sciences // View article

Towards a Decision Support System for Automated Selection of Optimal Neural Network Instance for Research and Engineering

Authors:
Rok Tavčar
Jože Dedič
Drago Bokal
Andrej Žemva

Keywords: Neural Networks; Decision Support System; Knowledge base; Taxonomy; Multiple-Criteria Optimization Problem.

Abstract:
The success of any advanced computing method (ACM) depends as much on its excellence as it does on a) whether it is optimally deployed and b) if it matches the problem at hand. Neural Networks, which are ACMs of remarkable potential, receive severe penalties in both of the latter aspects, due to reasons, put forth and addressed herein. This paper presents a theoretical foundation for an inference engine decision space and a taxonomic framework for a knowledge base, which are part of our proposed knowledge-driven decision support system (DSS) for optimal matching of a neural network (NN) setup against the given learning task. Such DSS supports solving a multiple criteria optimization problem, considering specific design constraints of the given NN-based machine learning application.

Pages: 78 to 85

Copyright: Copyright (c) IARIA, 2014

Publication date: August 24, 2014

Published in: conference

ISSN: 2308-4499

ISBN: 978-1-61208-354-4

Location: Rome, Italy

Dates: from August 24, 2014 to August 28, 2014