event
Thursday 07 Feb 2013: Kohonen Self Organising Maps for water resources data mining applications
Dr. Steve Mounce - University of Sheffield
Harrison 102 14:00-15:00
Dr. Steve Mounce is a Research Fellow in the Pennine Water Group (PWG), Department of Civil and Structural Engineering at the University of Sheffield. His PhD is in Computer Science and he has had nearly fifteen years’ experience on EPSRC, European and industrial sponsored research projects. His research combines Artificial Intelligence (particularly Artificial Neural Networks) and Water Engineering for hydroinformatics applications such as for leakage, event detection systems, data mining, ontologies and knowledge management. He won the IWEX University Challenge 2010 (Presented at Sustainabilitylive!) for an Artificial Intelligence burst detection system running on live data piloted at a UK water company. He is currently leading the Computer Science component of the EPSRC Pipe Dreams project, integrating across a multidisciplinary team which includes microbiologists.
A range of data driven tools have been applied in both hydroinformatics and bioinformatics for exploring the interrelationships between various types of variables, with a number of studies successfully using Artificial Neural Networks (ANNs) to probe complex data sets. Self Organising Maps (SOMs) are a class of ANN that perform dimensionality reduction of the feature space to yield topologically ordered maps. The SOM is trained without classes attached in an unsupervised fashion. Training combines competitive learning (learning the position of a data cloud) and co-operative learning (neighbours on map are adjusted to let it self-organise).
In this seminar, Steve describes a number of SOM applications from the water resources domain including their use in clustering microbiological T-RFLP data (for pilot facility water system experiments and for sewerage data from the field) and in data mining corporate databases. He will also summarise some other PWG research using AI techniques such as Particle Swarm Optimisation and Fuzzy Systems. Finally, he will indicate some current avenues of research and project collaborations with a Computer Science aspect.
