Abstract
Semi-supervised clustering based on pairwise constraints has been very active in recent years. The pairwise constraints consist of must-link and cannot-link. Since different types of constraints provide different information, they should be utilized with different strategies in the learning process. In this paper, we investigate the effect of must-link and cannot-link constraints on non-negative matrix factorization (NMF) and show that they play different roles when guiding the factorization procedure. A new semi-supervised NMF model is then proposed with pairwise constraints penalties. Among them, must-link constraints are used to control the distance of the data in the compressed form, and cannot-link constraints are used to control the encoding factor. Meanwhile, the same penalty strategies are applied on symmetric NMF model to handle the similarity matrix. The proposed two models are implemented by an alternating nonnegative least squares algorithm. We examine the performance of our models on series of real similarity data, and compare them with state-of-the-art, illustrating that the new models provide superior clustering performance.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Zhu, X., Lafferty, J., Ghahramani, Z.: Combining active learning and semi-supervised learning using gaussian fields and harmonic functions. In: Proc. of ICML (2003)
Chapelle, O., Scholkopf, B., Zien, A.: Semi-Supervised Learning. MIT Press (2006)
Pedrycz, W.: Knowledge-Based Clustering: From Data to Information Granules. John Wiley & Sons, Inc. (2005)
Yao, J.: A ten-year review of granular computing. In: Proc. of IEEE ICGC (2007)
Jing, L., Yu, J.: Text clustering based on granular computing and wikipedia. In: Proc. of WI-IAT (2010)
Wagstaff, K., Cardie, C., Rogers, S., Schroedl, S.: Constrained k-means clustering with background knowledge. In: Proc. of ICML, pp. 798–803 (2001)
Basu, S., Bilenko, M., Mooney, R.: A probabilistic framework for semi-supervised clustering. In: Proc. of ACM SIGKDD, pp. 59–68 (2004)
Hong, Y., Kwong, S.: Learning assignment order of instances for the constrained k-means clustering algorithm. IEEE Trans. Syst. Man Cybern. Part B 39, 568–574 (2009)
Klein, D., Kamvar, S., Manning, C.: From instance-level constraints to space-level constraints: making the most of prior knowledge in data clustering. In: Proc. of ICML (2002)
Shental, N., Bar-Hillel, A., Hertz, T., Weinshall, D.: Computing gaussian mixture models with em using equivalence constraints. In: Proc. of NIPS (2003)
Lu, Z., Leen, T.: Semi-supervised learning with penalized probabilistic clustering. In: Proc. of NIPS, vol. 17, pp. 849–856 (2005)
Coleman, T., Saunderson, J., Wirth, A.: Spectral clustering with inconsistent advice. In: Proc. of ICML, pp. 152–159 (2008)
Li, Z., Liu, J., Tang, X.: Constrained clustering via spectral regularization. In: Proc. of CVPR (2009)
Wang, X., Davidson, I.: Flexible constrained spectral clustering. In: Proc. of ACM SIGKDD, pp. 563–571 (2010)
Givoni, I., Frey, B.: Semi-supervised affinity propagation with instance-level constraints. In: Proc. of AISTATS (2009)
Li, T., Ding, C., Jordan, M.: Solving consensus and semi-supervised clustering problems using nonnegative matrix factorization. In: Proceedings of ICDM, pp. 577–582 (2007)
Chen, Y., Rege, M., Dong, M., Hua, J.: Non-negative matrix factorization for semi-supervised data clustering. Knowledge and Information Systems 17, 355–379 (2008)
Wang, F., Li, T., Zhang, C.: Semi-supervised clustering via matrix factorization. In: Proceedings of SIAM DM (2008)
Chen, Y., Wang, L., Dong, M.: Semi-supervised Document Clustering with Simultaneous Text Representation and Categorization. In: Buntine, W., Grobelnik, M., Mladenić, D., Shawe-Taylor, J. (eds.) ECML PKDD 2009, Part I. LNCS, vol. 5781, pp. 211–226. Springer, Heidelberg (2009)
Chen, Y., Wang, L., Dong, M.: Non-negative matrix factorization for semisupervised heterogeneous data coclustering. IEEE Trans. on Knowledge and Data Engineering 22, 1459–1474 (2010)
Liu, H., Wu, Z.: Non-negative matrix factorization with constraints. In: Proceedings of AAAI, pp. 506–511 (2010)
Lee, D., Seung, H.: Learning the parts of objects by nonnegative matrix factorization. Nature 401, 788–791 (1999)
Xu, W., Liu, X., Gong, Y.: Document clustering based on non-negative matrix factorization. In: Proc. of ACM SIGIR, pp. 267–273 (2003)
Shahnaz, F., Berry, M., Pauca, V., Plemmons, R.: Document clustering using nonnegative matrix factorization. Information Processing and Management 42, 373–386 (2006)
Berry, M., Gillis, N., Gilineur, F.: Document classification using nonnegative matrix factorization and underapproximation. In: Proc. of IEEE Intl. Symp. on Circuits and Systems, pp. 2782–2785 (2009)
Wang, F., Li, T., Wang, X., Zhu, S., Ding, C.: Community discovery using nonnegative matrix factorization. Data Mining Knowledge Discovery 22, 493–521 (2011)
Ding, C., He, X., Simon, H.: On the equivalence of nonnegative matrix factorization and spectral clustering. In: Proceedings of SIAM DM (2005)
Zhu, Y., Jing, L., Yu, J.: Text clustering with constrained non-negative matrix factorization. In: Proc. of ICDM, pp. 1278–1283 (2011)
Paatero, P.: The multilinear engine: A table-driven, least squares program for solving multilinear problems, including the n-way parallel factor analysis model. Journal of Computational and Graphical Statistics 8, 854–888 (1999)
Ding, C., Li, T., Peng, W., Park, H.: Orthogonal nonnegative matrix tri-factorizations for clustering. In: Proceedings of ACM SIGKDD (2006)
Lin, C.: Projected gradient methods for nonnegative matrix factorization. Neural Computation 19, 2756–2779 (2007)
Zhao, Y., Karypis, G.: Comparison of agglomerative and partitional document clustering algorithms. University of Minnesota. Tech. Rep. 02-014 (2002)
Chen, Y., Garcia, E., Gupta, M., Rahimi, A., Cazzanti, L.: Similarity-based classification: concepts and algorithms. J. of Machine Learning Research 10, 747–776 (2009)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2012 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Jing, L., Yu, J., Zeng, T., Zhu, Y. (2012). Semi-supervised Clustering via Constrained Symmetric Non-negative Matrix Factorization. In: Zanzotto, F.M., Tsumoto, S., Taatgen, N., Yao, Y. (eds) Brain Informatics. BI 2012. Lecture Notes in Computer Science(), vol 7670. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-35139-6_29
Download citation
DOI: https://doi.org/10.1007/978-3-642-35139-6_29
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-35138-9
Online ISBN: 978-3-642-35139-6
eBook Packages: Computer ScienceComputer Science (R0)