Applied neural networks
I Jagielska
6 points
* 2 hours per week
* First semester
* Caulfield
Objectives At the completion of this subject students should have in-depth knowledge of selected neural network architectures their theory and applications; be able to develop neural network applications for real-life business and industrial problems; and appreciate the current developments in neural networks.
Synopsis Neural computing/neural networks/parallel distributed processing is a fundamentally new approach to information processing inspired by research into the neural structure of the brain. Neural networks are renowned for their ability to learn and generalise from noisy and incomplete information. Neural networks are especially useful for unstructed problems because unlike conventional methods they do not require specification of the domain model they `learn' the underlying model from examples. This subject addresses the theory and applications of neural networks. There will be a strong emphasis on the development of neural network applications for business and industry. The following topics will be covered: the history, architecture and biological motivations, learning in neural networks, selected neural network architectures, applications of neural networks including forecasting, financial risk analysis, marketing, data mining and decision support; fuzzy neural networks, extracting rules from neural networks, development of neural network applications, current research in neural computing. The students will develop real-life neural network applications using packaged neural network software
Assessment Practical work: 60%
* Seminar paper (2500 words):
40%
| Published by Monash University, Clayton, Victoria
3168 Copyright © Monash University 1996 - All Rights Reserved - Caution Authorised by the Academic Registrar December 1996 |