Title :
Feature selection based on a black hole model of data reorganization
Author :
Markou, Markos ; Singh, Sameer
Author_Institution :
Dept. Of Comput. Sci., Exeter Univ., UK
Abstract :
This work introduces a new model of feature selection based on a pattern recognition model using the concept of black holes. We show that this method of feature selection is robust and provides an efficient subset of features for classification.
Keywords :
feature extraction; pattern classification; black hole model; data reorganization; feature selection; pattern recognition model; Clustering algorithms; Computer science; Covariance matrix; Extraterrestrial measurements; Feature extraction; Iterative algorithms; Pattern recognition; Robustness; Testing; Training data;
Conference_Titel :
Pattern Recognition, 2004. ICPR 2004. Proceedings of the 17th International Conference on
Print_ISBN :
0-7695-2128-2
DOI :
10.1109/ICPR.2004.1333835
