Recently, support vector machine (SVM) has become a popular tool in pattern recognition. In developing a successful SVM classifier, the first step is feature extraction. This paper proposes the application of independent component analysis (ICA) to SVM for feature extraction. In ICA, the original inputs are linearly transformed into features which are mutually statistically independent. By examining the Statlog heart disease data and satimage data, the experimental shows that SVM by feature extraction using ICA can perform better than that without feature extraction.
Combining independent component analysis with support vector machines
2006-01-01
305017 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Face Recognition Using Modular Principal Component Analysis and Support Vector Machines
British Library Conference Proceedings | 2006
|Face Recognition with Support Vector Machines: Global versus Component-based Approach
British Library Conference Proceedings | 2001
|Support vector machines for face recognition
British Library Online Contents | 2001
|