Abstract
Canonical Correlation Analysis (CCA) aims at identifying linear dependencies between two different but related multivariate views of the same underlying semantics. Ignoring its various extensions to more than two views, CCA uses these two views as complex labels to guide the search of maximally correlated projection vectors (covariates). Therefore, CCA can overfit the training data, meaning that different correlated projections can be found when the two-view training dataset is resampled. Although, to avoid such overfitting, ensemble approaches that utilize resampling techniques have been effectively used for improving generalization of many machine learning methods, an ensemble approach has not yet been formulated for CCA. In this paper, we propose an ensemble method for obtaining a final set of covariates by combining multiple sets of covariates extracted from subsamples. In comparison to those obtained by the application of the classical CCA on the whole set of training data, combining covariates with weaker correlations extracted from a number of subsamples of the training data produces stronger correlations that generalize to unseen test examples. Experimental results on emotion recognition, digit recognition, content-based retrieval, and multiple view object recognition have shown that ensemble CCA has better generalization for both the test set correlations of the covariates and the test set accuracy of classification performed on these covariates.
Similar content being viewed by others
References
Hotelling H (1936) Relations between two sets of variates. Biometrika 28:312–377
Hardoon DR, Szedmak S, Shawe-Taylor J (2004) Canonical correlation analysis: an overview with application to learning methods. Neural Comput 16:2639–2664
Kettenring JR (1971) Canonical analysis of several sets of variables. Biometrika 58:433–451
Kursun O, Alpaydin E, Favorov OV (2011) Canonical correlation analysis using within-class coupling. Pattern Recognit Lett 32(2):134–144
Blaschko MB, Shelton JA, Bartels A, Lampert CH, Gretton A (2011) Semi-supervised kernel canonical correlation analysis with application to human fMRI. Pattern Recognit Lett 32(11):1572–1583
Hardoon DR, Ettinger U, Mourão-Miranda J, Antonova E, Collier D, Kumari V, Williams SCR, Brammer M (2009) Correlation-based multivariate analysis of genetic influence on brain ume. Neurosci Lett 450:281–286
Waaijenborg S, Zwinderman AH (2009) Correlating multiple SNPs and multiple disease phenotypes: penalized non-linear canonical correlation analysis. Bioinformatics 25(21):2764–2771
Tripathi A, Klami A, Kaski S (2008) Simple integrative preprocessing preserves what is shared in data sources. BMC Bioinform 9(111). doi:10.1186/1471-2105-9-111
Gumus E, Kursun O, Sertbas A, Ustek D (2012) Application of canonical correlation analysis for identifying viral integration preferences. Bioinformatics 28(5):651–655
Tsay RS, Ling S (2008) Canonical correlation analysis for the vector AR(1) model with ARCH innovations. J Stat Plan Inference 138(9):2826–2836
Mishra SK (2009) Representation-constrained canonical correlation analysis: a hybridization of canonical correlation and principal component analyses. J Appl Econ Sci IV(7):115–124
Romanazzi M (1992) Influence in canonical correlation analysis. Psychometrika 57:237–259
Branco JA, Croux C, Filzmoser P, Oliveira MR (2005) Robust canonical correlations: a comparative study. Comput Stat 20:203–229
Sakar CO, Kursun O (2012) A method for combining mutual information and canonical correlation analysis: predictive mutual information and its use in feature selection. Expert Syst Appl 39(3):3333–3344
Lai PL, Fyfe C (1999) A neural implementation of canonical correlation analysis. Neural Netw 12(10):1391–1397
Hsieh WW (2000) Nonlinear canonical correlation analysis by neural networks. Neural Netw 13(10):1095–1105
Via J, Santamaria I, Perez J (2007) A learning algorithm for adaptive canonical correlation analysis of several data sets. Neural Netw 20:139–152
Karnel G (1991) Robust canonical correlation and correspondence analysis. In: The frontiers of statistical scientific and industrial applications. Proceedings of ICOSCO-I, The first international conference on statistical computing, vol II. American Sciences Press, Strassbourg, pp 335–354
Croux C, Dehon C (2002) Analyse canonique basee sur des estimateurs robustes de la matrice de covariance. Rev Stat Appl 2:5–26
Taskinen S, Croux C, Kankainen A, Ollila E, Oja H (2006) Canonical analysis based on scatter matrices. J Multivar Anal 97(2):359–384
Archambeau C, Delannay N, Verleysen M (2006) Robust probabilistic projections. In: Cohen W, Moore A (eds) Proceedings of ICML 2006, The 23rd international conference on machine learning. ACM, New York, pp 33–40
Viinikanoja J, Klami A, Kaski S (2010) Variational Bayesian mixture of robust CCA models, ECML PKDD 2010. Springer, Berlin, pp 370–385
Bach FR, Jordan MI (2005) A probabilistic interpretation of canonical correlation analysis. Tech Rep 688, Department of Statistics, University of California, Berkeley
Alpaydın E (2004) Introduction to machine learning. MIT Press, Cambridge, pp 351–371
Hansen L, Salamon P (1990) Neural network ensembles. IEEE Trans Pattern Anal Mach Intell 12:993–1001
Breiman L (1996) Bagging predictors. Mach Learn 24(2):123–140
Sharkey A (1999) Combining artificial neural nets. Springer, Berlin
Ghosh J (2002) Multiclassier systems: back to the future, (invited paper). In: Roli F, Kittler J (eds) Multiple classier systems. Springer, Berlin, pp 1–15
Okun O, Priisalu H (2005) Multiple views in ensembles of nearest neighbor classifiers. In: Proceedings of the workshop on learning with multiple views, 22nd ICML, Bonn, Germany
Lee H, Kim E, Pedrycz W (2013) A new selective neural network ensemble with negative correlation. Appl Intell 37(4):488–498. doi:10.1007/s10489-012-0342-3
Tay WL, Chui CK, Ong SH, Ng ACM (2013) Ensemble-based regression analysis of multimodal medical data for osteopenia diagnosis. Expert Syst Appl 40(2):811–819. doi:10.1016/j.eswa.2012.08.031
Siwek K, Osowski S (2012) Improving the accuracy of prediction of PM10 pollution by the wavelet transformation and an ensemble of neural predictors. Eng Appl Artif Intell 25(6):1246–1258. doi:10.1016/j.engappai.2011.10.013
Khor KC, Ting CY, Phon-Amnuaisuk S (2012) A cascaded classifier approach for improving detection rates on rare attack categories in network intrusion detection. Appl Intell 36(2):320–329
Garcia-Nieto J, Alba E (2012) Parallel multi-swarm optimizer for gene selection in DNA microarrays. Appl Intell 37(2):255–266
Strehl A, Ghosh J (2002) Cluster ensembles—a knowledge reuse framework for combining multiple partitions. J Mach Learn Res 3:583–617
Masoud H, Jalili S, Hasheminejad SMH (2013) Dynamic clustering using combinatorial particle swarm optimization. Appl Intell 38(3):289–314. doi:10.1007/s10489-012-0373-9
Ayad HG, Kamel MS (2010) On voting-based consensus of cluster ensembles. Pattern Recognit 43(5):1943–1953
Lu Z, Peng Y, Horace HS (2011) Combining multiple clusterings using fast simulated annealing. Pattern Recognit Lett 32(15):1956–1961
Mimaroglu S, Erdil E (2011) Combining multiple clusterings using similarity graph. Pattern Recognit 44:694–703
Zhang J, Zhang D (2011) A novel ensemble construction method for multi-view data using random cross-view correlation between within-class examples. Pattern Recognit 44:1162–1171
Mo KC, Thiaw WM (2002) Ensemble canonical correlation prediction of precipitation over the Sahel. Geophys Res Lett 29(12):4
Shao J, Tu D (1995) The jackknife and bootstrap. Springer, Berlin
Lucey P, Cohn JF, Kanade T, Saragih J, Ambadar Z, Matthews I (2010) The extended Cohn-Kande dataset (CK+): a complete facial expression dataset for action unit and emotion-specified expression. In: Third IEEE workshop on CVPR for human communicative behavior analysis
Karaali A (2012) Face detection and facial expression recognition using moment invariants. MSc Thesis, Bahcesehir University
Sakar CO, Kursun O, Karaali A, Erdem CE (2012) Feature extraction for facial expression recognition by canonical correlation analysis. In: IEEE 20th signal processing and applications conference (SIU), Mugla, Turkey
Ulukaya S (2011) Affect recognition from facial expressions for human–computer interaction. MSc Thesis, Bahcesehir University
Asuncion A, Newman DJ (2007) UCI machine learning repository. University of California, Department of Information and Computer Science, Irvine, CA
Xing EP, Yan R, Hauptmann AG (2005) Mining associated text and images with dual-wing harmoniums. In: Conference on uncertainty in artificial intelligence, pp 633–641
Nene S, Nayar S, Murase H (1996) Columbia Object Image Library (COIL-100). Technical Report CUCS-006-96, New York: Columbia Univ, Press
Hsu CW, Lin CJ (2002) A comparison of methods for multi-class support vector machines. IEEE Trans Neural Netw 13:415–425
Acknowledgements
The work of C.O. Sakar is supported by the Ph.D. scholarship (2211) from Turkish Scientific Technical Research Council (TÜBİTAK). He is a Ph.D. student in the Computer Engineering Department at Bogazici University, Istanbul, Turkey.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Sakar, C.O., Kursun, O. & Gurgen, F. Ensemble canonical correlation analysis. Appl Intell 40, 291–304 (2014). https://doi.org/10.1007/s10489-013-0464-2
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10489-013-0464-2