access icon free Non-negative matrix factorisation based on fuzzy K nearest neighbour graph and its applications

Non-negative matrix factorisation (NMF) has been widely used in pattern recognition problems. For the tasks of classification, however, most of the existing variants of NMF ignore both the discriminative information and the local geometry of data into the factorisation. The actual conditions of the problems will be affected by the change of the environmental factors to affect the recognition accuracy. In order to overcome these drawbacks, the authors regularised NMF by intra-class and inter-class fuzzy K nearest neighbour graphs, leading to NMF-FK-NN in this study. By introducing two novel fuzzy K nearest neighbour graphs, NMF-FK-NN can contract the intra-class neighbourhoods and expand the inter-class neighbourhoods in the decomposition. This method not only exploits the discriminative information and uses the geometric structure in the data effectively, but also reduces the influence of the external factors to improve recognition effect. In the factorisation, the authors minimised the approximation error whilst contracting intra-class fuzzy neighbourhoods and expanding inter-class fuzzy neighbourhoods. The authors develop simple multiplicative updates for NMF-FK-NN and present monotonic convergence results. Experiments of the text clustering on the CLUTO toolkit and face recognition on ORL and YALE datasets show the effectiveness of our proposed method.

Inspec keywords: face recognition; graph theory; matrix decomposition; pattern recognition

Other keywords: approximation error; pattern recognition problems; inter-class fuzzy neighbourhoods; CLUTO toolkit; monotonic convergence; YALE datasets; intra-class fuzzy K nearest neighbour graphs; NMF-FK-NN; fuzzy K nearest neighbour graph; nonnegative matrix factorisation; fuzzy K nearest neighbour graphs; geometric structure; data representation method; ORL datasets; inter-class fuzzy K nearest neighbour graphs; intra-class fuzzy neighbourhoods

Subjects: Combinatorial mathematics; Combinatorial mathematics; Algebra; Computer vision and image processing techniques; Image recognition; Algebra

References

    1. 1)
      • 3. Lee, D.D., Seung, H.S.: ‘Learning the parts of objects by nonnegative matrix factorization’, Nature, 1999, 401, (21), pp. 788791.
    2. 2)
      • 8. Li, S.Z., Hou, X., Zhang, H., Cheng, Q.: ‘Learning spatially localized, parts-based representation’. Proc. IEEE Int. Conf. on Comput. Vis. Pattern Recognition, 2001, pp. 207212.
    3. 3)
      • 14. Keller, J.M., Gray, M.R., Givens, J.A.: ‘A fuzzy k-nearest neighbor algorithm’, IEEE Trans. Syst., Man Cybernet., 1985, 15, (4), pp. 580585.
    4. 4)
      • 17. Lee, D.D., Seung, H.S.: ‘Algorithms for nonnegative matrix factorization’, Adv. Neural Inf. Process. Syst. (NIPS), 2000, pp. 556562.
    5. 5)
      • 11. Zafeiriou, S., Tefas, A., Buciu, I., Pitas, I.: ‘Exploiting discriminant information in nonnegative matrix factorization with application to frontal face verification’, IEEE Trans. Neural Netw., 2006, 17, (3), pp. 683695 (doi: 10.1109/TNN.2006.873291).
    6. 6)
      • 10. Wang, Y., Jia, Y., Hu, C., Turk, M.: ‘Fisher non-negative matrix factorization for learning local features’. Proc. Asian Conf. on Computer Vision (ACCV), Jeju Island, Korea, 2004.
    7. 7)
      • 5. Belkin, M., Niyogi, P.: ‘Laplacian eigenmaps and spectral techniques for embedding and clustering’, Adv. Neural Inf. Process. Syst., 2002, 14, pp. 585591.
    8. 8)
      • 4. Cai, D., He, X., Wu, X., Han, J.: ‘Non-negative matrix factorization on manifold’. Proc. IEEE Int. Conf. on Data Mining, 2008, pp. 6372.
    9. 9)
      • 1. Kirby, M., Sirovich, L.: ‘Application of the Karhunen-Loeve procedure for the characterization of human faces’, IEEE Trans. PAMI, 1990, 12, (1), pp. 103108 (doi: 10.1109/34.41390).
    10. 10)
      • 18. Lovasz, L.: ‘Matching Theory (North-Holland Mathematics Studies)’ (Elsevier Science Ltd., 1986).
    11. 11)
      • 7. Guan, N., Tao, D., Luo, Z., Yuan, B.: ‘Manifold regularized discriminative nonnegative matrix factorization with fast gradient descent’, IEEE Trans. Image Process., 2011, 20, (7), pp. 20302048 (doi: 10.1109/TIP.2011.2105496).
    12. 12)
      • 6. He, X., Niyogi, P.: ‘Locality preserving projections’, Adv. Neural Inf. Process. Syst., 2003, 15, pp. 18.
    13. 13)
      • 12. An, S., Yoo, J., Choi, S.: ‘Manifold-respecting discriminant nonnegative matrix factorization’, Pattern Recognit. Lett., 2011, 32, pp. 832837 (doi: 10.1016/j.patrec.2011.01.012).
    14. 14)
      • 9. Hoyer, P.: ‘Non-negative matrix factorization with sparseness constraints’, J. Mach. Learn. Res., 2004, 5, pp. 14571469.
    15. 15)
      • 15. Wan, M., Yang, G., Lai, Z., Jin, Z.: ‘Feature extraction based on fuzzy local discriminant embedding with applications to face recognition’, IET Comput. Vis., 2011, 5, (5), pp. 301308 (doi: 10.1049/iet-cvi.2011.0028).
    16. 16)
      • 13. Kwak, K.C., Pedrycz, W.: ‘Face recognition using a fuzzy Fisherface classifier’, Pattern Recognit., 2005, 38, (10), pp. 17171732 (doi: 10.1016/j.patcog.2005.01.018).
    17. 17)
      • 16. Chen, H.T., Chang, H.W., Liu, T.L.: ‘Local discriminant embedding and its variants’. Proc. IEEE Computer Society Conf. Computer Vision and Pattern Recognition, 2005, pp. 846853.
    18. 18)
      • 2. Chengjun, L., Wechsler, H.: ‘Independent component analysis of Gabor features for face recognition’, IEEE Trans. Neural Netw., 2003, 14, (4), pp. 919928 (doi: 10.1109/TNN.2003.813829).
http://iet.metastore.ingenta.com/content/journals/10.1049/iet-cvi.2013.0055
Loading

Related content

content/journals/10.1049/iet-cvi.2013.0055
pub_keyword,iet_inspecKeyword,pub_concept
6
6
Loading