Canonical correlation analysis for multiview semisupervised feature extraction

Authors:
Olcay Kursun;Ethem Alpaydin
Affiliations:
Department of Computer Engineering, Istanbul University, Avcilar, Istanbul, Turkey;Department of Computer Engineering, Bogazici University, Bebek, Istanbul, Turkey
Venue:
ICAISC'10 Proceedings of the 10th international conference on Artificial intelligence and soft computing: Part I
Year:
2010

Citing 5
Cited 2

SINBAD: A neocortical mechanism for discovering environmental variables and regularities hidden in sensory input

Biological Cybernetics
Introduction to Machine Learning (Adaptive Computation and Machine Learning)

Introduction to Machine Learning (Adaptive Computation and Machine Learning)
Canonical Correlation Analysis: An Overview with Application to Learning Methods

Neural Computation
Dimensionality reduction of image features using the canonical contextual correlation projection

Pattern Recognition
A comparison of methods for multiclass support vector machines

IEEE Transactions on Neural Networks

Canonical correlation analysis using within-class coupling

Pattern Recognition Letters
A unified dimensionality reduction framework for semi-paired and semi-supervised multi-view data

Pattern Recognition

Quantified Score

Hi-index	0.00

Visualization

Abstract

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.