Canonical Correlation Analysis for Multiview Semisupervised Feature Extraction
Hotelling’s Canonical Correlation Analysis (CCA) works with two sets of related variables, also called views, and its goal is to find their linear projections with maximal mutual correlation. CCA is most suitable for unsupervised feature extraction when given two views but it has been also long known that in supervised learning when there is only a single view of data given, the supervision signal (class-labels) can be given to CCA as the second view and CCA simply reduces to Fisher’s Linear Discriminant Analysis (LDA). However, it is unclear how to use this equivalence for extracting features from multiview data in semisupervised setting (i.e. what modification to the CCA mechanism could incorporate the class-labels along with the two views of the data when labels of some samples are unknown). In this paper, a CCA-based method supplemented by the essence of LDA is proposed for semi-supervised feature extraction from multiview data.
KeywordsSemisupervised Learning Feature Extraction Multiview Learning LDA CCA
Unable to display preview. Download preview PDF.
- 1.Fisher, R.A.: The Use of Multiple Measurements in Taxonomic Problems. Annals of Eugenics 7, 179–188 (1936)Google Scholar
- 7.Alpaydin, E.: Introduction to Machine Learning (Adaptive Computation and Machine Learning Series). The MIT Press, Cambridge (2004)Google Scholar
- 11.van Breukelen, M., Duin, R.P.W., Tax, D.M.J., den Hartog, J.E.: Handwritten digit recognition by combined classifiers. Kybernetika 34(4), 381–386 (1998)Google Scholar
- 12.Asuncion, A., Newman, D.J.: UCI Machine Learning Repository. University of California, Department of Information and Computer Science, Irvine (2007)Google Scholar
- 13.Borga, M.: Learning Multidimensional signal processing, PhD thesis, Department of Electrical Engineering, Linköping University, Linköping, Sweden (1998)Google Scholar