Abstract
In this paper, an unsupervised multi-manifold Isomap algorithm, which is named UMD-Isomap, is proposed for the purpose of dimensionality reduction and clustering of multi-manifold data. First, the global pairwise constraints are constructed by training m mixtures of probabilistic principal component analyzers (MPPCA) and propagating their local tangent subspaces. At the same time, the sub-manifolds are also clustered, and their classes information are recorded in the pairwise constraints. If the number of sub-manifolds is known, a new pairwise constraints is computed by using a cluster ensemble algorithm, which creates a similarity matrix by accumulating c sets of pairwise constraints. Subsequently, a new objective function with pairwise constraints and two supervised solutions are proposed to achieve the dimensionality reduction of the multi-manifolds. The proposed UMD-Isomap algorithm achieved better performance in terms of dimensionality reduction and clustering accuracy than other commonly used methods and its effectiveness was verified.
Similar content being viewed by others
Notes
The experiment is carried out with the python version of Umap algorithm, and its results in figures are displayed by MATLAB.
References
Tenenbaum J, Silva DD, Langford J (2000) A global geometric framework for nonlinear dimensionality reduction. Science 290(5500):2319–2323
Roweis S, Saul L (2000) Nonlinear dimentionality reduction by locally linear embedding. Science 290(5500):2323–2326
Belkin M, Niyogi P (2003) Laplacian eigenmaps for dimensionality reduction and data representation. Neural Comput 15(6):1373–1396
Lin T, Zha HB (2008) Riemannian manifold learning. IEEE Trans. Pattern Anal. Mach. Intell. 30(5):796–809
Liu Y, Nie F et al (2019) Flexible unsupervised feature extraction for image classification. Neural Netw. 115:65–71
Liu Y, Gao Q, Yang Z et al (2018) Learning with Adaptive Neighbors for Image Clustering. In: Twenty-Seventh International Joint Conference on Artificial Intelligence IJCAI-18. 2018, 2483-2489
Fan B, Kong QQ et al (2020) Efficient nearest neighbor search in high dimensional hamming space. pattern recognition. https://doi.org/10.1016/j.patcog.2019.107082
Zheng Y, Liu X, Chen S et al (2020) Multi-task deep dual correlation filters for visual tracking. IEEE Trans. Image Process. 29:9614–9626
Zheng YH, Jeon B et al (2018) Student’s t-Hidden Markov Model for Unsupervised Learning Using Locallized Feature Selection. IEEE Transactions on Circuits and Systems for Video Technology 28(10):2586–2598
Souvenir R, Manifold Pless R (2005) Clustering. Proceedings 1(1):648–653
Meng D, Leung Y, Fung T, Xu Z (2008) Nonlinear dimensionality reduction of data lying on the multicluster manifold. IEEE Trans Syst Man Cybernet-Part B 38(4):1111–1122
Gao XF, Liang JY (2013) Manifold learning algorithm DC-ISOMAP of data lying on well-separated multi-manifold with same intrinsic dimension. J Comput Res Dev 50(8):1690–1699
Ng A, Jordan M, Weiss Y (2001) On spectral clustering: Analysis and an algorithm. in Advances in Neural Information Processing Systems 14. 2001: 849-856
Shi JB, Malik J (2000) Normalized cuts and image segmentation. IEEE Trans Pattern Anal Mach Intell 22(8):888–905
Zhang K, Kwok JT (2010) Clustered Nyström method for large scale manifold learning and dimension reduction. IEEE Trans Neural Netw 21(10):1576–1587
Wang Y, Jiang Y, Zhou ZH (2011) Spectral clustering on multiple manifolds. IEEE Trans Neural Netw 22(7):1149–1161
Wang Y, Jiang Y, Wu Y, et al (2011) Local and Structural Consistency for Multi-manifold Clustering. Proc of IJCAI 2011. Menlo Park, CA: AAAI Press, 1559-1564
Babaeian A, Bayestehtashk A, Bandarabadi M. Multiple Manifold Clustering Using Curvature Constrained Path. PLoS ONE 10(9): e0137986. https://doi.org/10.1371/journal.pone.0137986
van der Maaten L, Hinton GE (2008) Visualizing high-dimensional data using t-SNE. J Mach Learn Res 9:2579–2605
van der Maaten L (2014) Accelerating t-SNE using Tree-Based Algorithms. Journal of Machine Learning Research 15(Oct):3221-3245
McInnes L, Healy J, Melville J (2018) UMAP: uniform manifold approximation and projection for dimension reduction. J Open Source Softw 3(29):861
Basu S, Davidson I, Wagstaff K (2008) Constrained clustering: advanced in algorithms, theory, and applications. CRC Press, Boca Raton
Geng X, Dc Zhan, Zhou ZH (2005) supervised nonlinear dimensionality reduction for visualization and classification. IEEE Trans Syst Man Cybernet-Part B 35(6):1098–1107
Goldberg A, Zhu X et al (2009) Multi-manifold semi-supervised learning. In: Proceedings of the 12th International Conference on Artificial Intelligence and Statistics, 2009, 169-176
Zhang Z, Chow TW, Zhao M (2013) M-Isomap:orthogonal constrained marginal Isomap for nonlinear dimensionality reduction. IEEE Trans. Cybern 43(1):180–191
Yang B, Xiang M, Zhang Y (2016) Multi-manifold Discriminant Isomap for visualization and classification. Pattern Recognit 2016(55):215–230
Zhang Y, Zhang Z et al (2017) Semi-supervised local multi-manifold Isomap by linear embedding for feature extraction. Pattern Recoginit 2017(000):1–17
Fan B, Liu HM et al (2020) Deep Unsupervised Binary Descriptor Learning through Locality Consistency and Self Distinctiveness. IEEE Transactions on Multimedia. https://doi.org/10.1109/TMM.2020.3016122
Tipping ME, Bishop CM (1999) Mixtures of probabilistic principal component analysers. Neural Comput. 11(2):443–482
Zhao J (2014) Efficient Mofel selection for mixtures of probabilistic PCA via hierarchical BIC. IEEE Trans Cybernet 44(10):1871–1883
Dempster AP, Laird NM, Rubin DB (1977) Maximum likelihood from incomplete data via the EM algorithm. J R Stat Soc Ser B (Methodological) 39(1):1–38
Sun D, Zhang D (2010) Bagging constraint score for feature selection with pairwise constraints. Pattern Recognit 43(6):2106–2118
Baghshah M S , Shouraki S B. Semi-supervised metric learning using pairwise constraints. In: Proceedings of the International Joint Conference on Artificial Intelligence, 2009, 9: 1217-1222
Wang F (2011) Semi-supervised metric learning by maximizing constraint margin. IEEE Trans Syst, Man Cybernet Part B-Cybernet 41(4):931–939
Joint Optimization for Pairwise Constraint Propagation, IEEE TNNLS, 2020
(2020) Pairwise Constraint Propagation With Dual Adversarial Manifold Regularization. IEEE TNNLS
Leeuw J de. Applications of Convex Analysis to Multidimensional Scaling. In JR Barra, F Brodeau, G Romier, B van Cutsem (eds.), Recent Developments in Statistics (1977) 133–145. North Holland Publishing Company, Amsterdam
de Leeuw I, Mair P (2009) Multidimensional scaling using majorization: SMACOF in R. J Stat Softw 31(3):1–30
Saul L, Roweis S (2004) Think globally, fit locally: unsupervised learning of low dimensional manifolds. J Mach Learn Res 4(2):119–155
Donoho D, Grimes C (2003) Hessian eigenmaps: locally linear embedding techniques for high-dimensional data. PNAS. 100(10):5591–5596
Zhang ZY, Zha HY (2005) Principal manifolds and nonlinear dimensionality reduction via tangent space alignment. SIAM J Sci Comput 26(1):313–338
Gao XF, Liang JY (2011) The dynamical neighborhood selection based on the sampling density and manifold curvature for isometric data embedding. Pattern Recognit Lett 32:202–209
Steyvers M (2002) Multidimensional scaling, In Encyclopedia of. Cognitive Science
Acknowledgements
This work is partially supported by the National Natural Science Foundation of China (No. 61703252); Applied Basic Research Programs of Shanxi Province (201701D121053) and Research Project Supported by Shanxi Scholarship Council of China (2016-002).
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Gao, X., Liang, J., Wang, W. et al. An unsupervised multi-manifold discriminant isomap algorithm based on the pairwise constraints. Int. J. Mach. Learn. & Cyber. 13, 1317–1336 (2022). https://doi.org/10.1007/s13042-021-01449-8
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s13042-021-01449-8