Advertisement

Knowledge and Information Systems

, Volume 25, Issue 3, pp 607–622 | Cite as

Spectral clustering in multi-agent systems

  • Balint TakacsEmail author
  • Yiannis Demiris
Regular Paper

Abstract

We examine the application of spectral clustering for breaking up the behavior of a multi-agent system in space and time into smaller, independent elements. We propose clustering observations of individual entities in order to identify significant changes in the parameter space (like spatial position) and detect temporal alterations of behavior within the same framework. Available knowledge of important interactions (events) between entities is also considered. We describe a novel algorithm utilizing iterative subdivisions where clusters are pre-processed at each step to counter spatial scaling, rotation, replay speed, and varying sampling frequency. A method is presented to balance spatial and temporal segmentation based on the expected group size, and a validity measure is introduced to determine the optimal number of clusters. We demonstrate our results by analyzing the outcomes of computer games and compare our algorithm to K-means and traditional spectral clustering.

Keywords

Spectral clustering Spatio-temporal data mining Multi-agent systems Plan extraction Plan recognition 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Assent I, Krieger R, Glavic B, Seidl T (2008) Clustering multidimensional sequences in spatial and temporal databases. Knowl Inf Syst 16(1): 29–51. doi: 10.1007/s10115-007-0121-3 CrossRefGoogle Scholar
  2. 2.
    Bach FR, Jordan MI (2006) Learning spectral clustering, with application to speech separation. J Mach Learn Res 7: 1963–2001MathSciNetGoogle Scholar
  3. 3.
    Bengio Y, Delalleau O, Le Roux N, Paiement J-F, Vincent P, Ouimet M (2004) Learning eigenfunctions links spectral embedding and kernel PCA. Neural Comput 16(10): 2197–2219zbMATHCrossRefGoogle Scholar
  4. 4.
    Chan J, Bailey J, Leckie C (2008) Discovering correlated spatio-temporal changes in evolving graphs. Knowl Inf Syst 16(1): 53–96. doi: 10.1007/s10115-007-0117-z CrossRefGoogle Scholar
  5. 5.
    Fischer I, Poland J (2005) Amplifying the block matrix structure for spectral clustering. In: Proceedings of the 14th annual machine learning conference of Belgium and the Netherlands, pp 21–28Google Scholar
  6. 6.
    Goyal VK, Vetterli M (1998) Block transform adaptation by stochastic gradient descent. In: IEEE digital signal processing workshop, Bryce Canyon, UT, August, p 75Google Scholar
  7. 7.
    Ham J, Ahn I, Lee D (2006) Learning a manifold-constrained map between image sets: applications to matching and pose estimation. In: Proceedings of the 2006 IEEE Computer Society conference on computer vision and pattern recognition, vol 1. IEEE Computer Society, Washington, pp 817–824Google Scholar
  8. 8.
    Hmlinen W (2009) StatApriori: an efficient algorithm for searching statistically significant association rules. Knowl Inf Syst. doi: 10.1007/s10115-009-0229-8
  9. 9.
    Hubert L, Arabie P (1985) Comparing partitions. J Classification 2(1): 193–218CrossRefGoogle Scholar
  10. 10.
    Jenkins OC, Mataric MJ (2004) A spatio-temporal extension to isomap nonlinear dimension reduction. In: Proceedings of ICML, Banff, Canada, 4–8 July 2004. ACM, New York, pp 441–448Google Scholar
  11. 11.
    Lafon S, Keller Y, Coifman RR (2006) Data fusion and multicue data matching by diffusion maps. IEEE Trans Pattern Anal Mach Intell 28(11): 1784–1797CrossRefGoogle Scholar
  12. 12.
    Mahadevan S, Maggioni M (2006) Proto-value functions: a Laplacian framework for learning representation and control in Markov decision processes. Technical Report 2006-35, July 2006Google Scholar
  13. 13.
    Newman MEJ (2006) Modularity and community structure in networks. Proc Natl Acad Sci 103(23): 8577–8582CrossRefGoogle Scholar
  14. 14.
    Ng A, Jordan M, Weiss Y (2002) On spectral clustering: analysis and an algorithm. In: Advances in neural information processing systems 14. MIT Press, Cambridge, p 849Google Scholar
  15. 15.
    Papapetrou P, Kollios G, Sclaroff S, Gunopulos D (2009) Mining frequent arrangements of temporal intervals. Knowl Inf Syst. doi: 10.1007/s10115-009-0196-0
  16. 16.
    Park J, Zha H, Kasturi R (2004) Spectral clustering for robust motion segmentation. In: Lecture notes in computer science, pp 390–401Google Scholar
  17. 17.
    Porikli F (2005) Ambiguity detection by fusion and conformity: a spectral clustering approach. In: International conference on integration of knowledge intensive multi-agent systems, 18–21 April 2005, pp 366–372Google Scholar
  18. 18.
    Ruan J, Zhang W (2007) An efficient spectral algorithm for network community discovery and its applications to biological and social networks. In: Proceedings of IEEE ICDM-07, Omaha, NE, USA, 28–31 October 2007Google Scholar
  19. 19.
    Saul LK, Weinberger KQ, Ham JH, Sha F, Lee DD (2006) Semisupervised learning. In: Spectral methods for dimensionality reduction. MIT Press, CambridgeGoogle Scholar
  20. 20.
    Shi J, Malik J (2000) Normalized cuts and image segmentation. IEEE Trans Pattern Anal Mach Intell 22(8): 888–905CrossRefGoogle Scholar
  21. 21.
    Şimşek Ö, Wolfe AP, Barto AG (2005) Identifying useful subgoals in reinforcement learning by local graph partitioning. In: Proceedings of ICML, vol 119 of ACM international conference proceeding series, pp 816–823Google Scholar
  22. 22.
    Sugar CA, James GM (2003) Finding the number of clusters in a dataset. J Am Stat Assoc 98(463): 750–763zbMATHCrossRefMathSciNetGoogle Scholar
  23. 23.
    Yuan J, Zhang B, Lin F (2005) Graph partition model for robust temporal data segmentation. In: Ho TB, Cheung D, Liu H (eds) Proceedings of PAKDD, number 3518 in LNAI. Springer-Verlag, Berlin, pp 758–763Google Scholar
  24. 24.
    Zelnik-Manor L, Irani M (2006) Statistical analysis of dynamic actions. IEEE Trans Pattern Anal Mach Intell 28(9): 1530–1535CrossRefGoogle Scholar
  25. 25.
    Zelnik-Manor L, Perona P (2004) Self-tuning spectral clustering. Adv Neural Inf Process Syst 17(1601–1608): 16Google Scholar
  26. 26.
    Zivkovic Z, Bakker B, Krose B (2006) Hierarchical map building and planning based on graph partitioning. In: IEEE international conference on robotics and automation, pp 803–809Google Scholar

Copyright information

© Springer-Verlag London Limited 2009

Authors and Affiliations

  1. 1.Intelligent Systems and Networks Group, Electrical and Electronic EngineeringImperial CollegeLondonUK

Personalised recommendations