This seminar will look at the use of evolutionary computation techniques to improve the perfomance of k-nearest neighbour classification on a well known intrusion detection data set. Two genetic algorithms (real-coded and bitstring) are used to transform the features of the data set before classification takes place. Additionally, the methods applied can assign ranks to the features within the data set. This ranking can be used to perform feature selection of the data set, reducing it's dimensionality and improving overall classification performance.
Last modified: Friday, 16-Jul-2004 15:37:30 NZST
This page is maintained by Caroline Wills.