Title :
A New Approach to Classification with the Least Number of Features
Author :
Klement, Sascha ; Martinetz, Thomas
Author_Institution :
Inst. for Neuroand Bioinf., Univ. of Lubeck, Lubeck, Germany
Abstract :
Recently, the so-called Support Feature Machine (SFM) was proposed as a novel approach to feature selection for classification, based on minimisation of the zero norm of a separating hyper plane. We propose an extension for linearly non-separable datasets that allows a direct trade-off between the number of misclassified data points and the number of dimensions. Results on toy examples as well as real-world datasets demonstrate that this method is able to identify relevant features very effectively.
Keywords :
learning (artificial intelligence); minimisation; pattern classification; SFM; classification; feature selection; linearly nonseparable datasets; separating hyper plane; support feature machine; zero norm minimisation; Bioinformatics; Input variables; Machine learning; Minimization; Noise; Support vector machines; Training; Support feature machine; classification; feature selection; zero norm minimisation;
Conference_Titel :
Machine Learning and Applications (ICMLA), 2010 Ninth International Conference on
Conference_Location :
Washington, DC
Print_ISBN :
978-1-4244-9211-4
DOI :
10.1109/ICMLA.2010.28
