Self-organising Maps to Study the Effects of Urbanisation at Long Bay in New Zealand
| aut.conference.type | Paper Published in Proceedings | |
| aut.relation.endpage | 137 | |
| aut.relation.startpage | 132 | |
| aut.researcher | Shanmuganathan, Subana | |
| dc.contributor.author | Shanmuganathan, S | |
| dc.contributor.author | Sallis, P | |
| dc.contributor.author | Buckeridge, JS | |
| dc.date.accessioned | 2013-04-11T21:10:16Z | |
| dc.date.available | 2013-04-11T21:10:16Z | |
| dc.date.copyright | 2001 | |
| dc.date.issued | 2001 | |
| dc.description.abstract | Biologically inspired Artificial Neural Network (ANN) modelling methods provide a means of problem solving that incorporates heuristics with conventional algorithmic processing Over the last few decades, nell' techniques for neuron relationship modelling and network architecture algorithms have been introduced to solve a ll'ide range of problems across many fields. This paper looks into the aspects of using SOMs to a biological example to permit understanding of a complex environmental process. The preliminmy research results to study the effects of urbanisation on marine life at the Long Bay-Okura Marine Reserve, situated in northern Nell' Zealand is discussed in detail. This was the count1y 's first urban, marine resen1e to be established (1995), and resulted from groll'ing concern of environmental groups and general public of the area. Since then many institutions have conducted research to find the cause for the observed environmental change. All these data sets are fused and analysed collectively to study the patterns in them. The use of SOMs to industrial process monitoring has been ve1y successful in many areas and is applied here to study the process dynamics in environmental process modelling with SOM trajectories. The analyses show the relationships found in the data sets from different sources in easily perceivable formats without having to model the complex physical process. | |
| dc.identifier.citation | Fifth Biennial Conference on Artificial Neural Networks and Expert Systems, Dunedin, New Zealand, 2001-11-25 to 2001-11-29, pp.132 - 137 | |
| dc.identifier.uri | https://hdl.handle.net/10292/5269 | |
| dc.publisher | Knowledge Engineering Lab, the University of Otago | |
| dc.rights | All rights are reserved. These proceedings or any part thereof, may not be reproduced in any form without the expressed written permission of the Editors. However, selected material may be copied, with due acknowledgement, for the purpose of teaching and furthering knowledge in intelligent information systems. The Editors and the University of Otago do not accept responsibility for the opinions expressed in the papers. The opinions are those held by the authors, and the Editors and the University of Otago make no endorsements, implied or otherwise, of these opinions and views. | |
| dc.rights.accessrights | OpenAccess | |
| dc.subject | Self-organising maps | |
| dc.subject | Urbanisation | |
| dc.subject | Data mining | |
| dc.subject | Environmental process monitoring | |
| dc.title | Self-organising Maps to Study the Effects of Urbanisation at Long Bay in New Zealand | |
| dc.type | Conference Contribution | |
| pubs.elements-id | 9884 | |
| pubs.organisational-data | /AUT | |
| pubs.organisational-data | /AUT/Health & Environmental Science | |
| pubs.organisational-data | /AUT/Vice Chancellor's Group |