Studley, M. , Bagnall, A. J. , Bull, L. and Whittley, I. M.
Super Computer Heterogeneous Classifier Meta-Ensembles.
To be published in International Journal of Data Warehousing and Mining, 3 (2).
Full text not available from this repository
|Additional Information:||As the use of machine learning for exploratory data analysis has increased, so have the sizes of the datasets they must face and the sophistication of the algorithms themselves. For this reason there is a growing body of research concerned with the use of parallel computing for data mining.
The Supercomputer Data Mining (SCDM) Project (EPSRC project (GR/T18455/01)) produced a super-computing data mining resource for use by the UK academic community. In particular, a number of evolutionary computing-based algorithms and the ensemble machine approach were used to exploit the large-scale parallelism possible in super-computing.
In an ensemble machine many classifiers can be used to increase accuracy and speed of classification through parallelism. This paper presents results of using ensembles of heterogeneous classifier techniques, which may themselves be ensembles of classifiers.
The increased accuracy thus achieved would only be practical given a Supercomputer resource.
The techniques and implementation detailed herein are now in use in data mining UK Olympic training data, England cricket bowling data, cancer data, etc.|
|Faculty/Department:||Faculty of Environment and Technology|
~Pre-2012 Faculty Structure > Faculty of Environment and Technology
~Pre-2010 Faculty Structure > Environment and Technology > Bristol Institute of Technology
R. Upload account
|Deposited On:||22 Jan 2010 15:10|
|Last Modified:||22 Nov 2012 15:35|
Repository Staff Only: item control page