Advertisement

Cybernetics and Systems Analysis

, Volume 44, Issue 1, pp 125–138 | Cite as

On the time series support vector machine using dynamic time warping kernel for brain activity classification

  • W. A. Chaovalitwongse
  • P. M. Pardalos
Article

Abstract

A new data mining technique used to classify normal and pre-seizure electroencephalograms is proposed. The technique is based on a dynamic time warping kernel combined with support vector machines (SVMs). The experimental results show that the technique is superior to the standard SVM and improves the brain activity classification.

Keywords

time series classification EEG brain dynamics optimization dynamic time warping epilepsy support vector machines 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    A. Babloyantz and A. Destexhe, “Low dimensional chaos in an instance of epilepsy,” Proc. Nat. Acad. Sci. USA, 83, 3513–3517 (1986).CrossRefGoogle Scholar
  2. 2.
    D. Berndt and J. Clifford, “Using dynamic time warping to find patterns in time series,” in: Proc. of the AAAI-94 Workshop on Knowledge Discovery in Databases (KDD-94) (1994).Google Scholar
  3. 3.
    D. Bertsimas, C. R. Darnell, and R. Soucy, “Portfolio construction through mixed-integer programming at Grantham, Mayo, Van Otterloo and Company,” Interfaces, 29, No. 1, 49–66 (1999).Google Scholar
  4. 4.
    M. K. Borgwardt, S. V. N. Vishwanathan, and H-P. Kriegel, “Class prediction from time series gene expression profiles using dynamical systems kernels,” in: Pacific Symp. on Biocomput (2006), pp. 547–558.Google Scholar
  5. 5.
    P. S. Bradley, U. Fayyad, and O. L. Mangasarian, “Mathematical programming for data mining: Formulations and challenges,” INFORMS J. Computing, 11, 217–238 (1999).MATHMathSciNetCrossRefGoogle Scholar
  6. 6.
    P. S. Bradley, O. L. Mangasarian, and W. N. Street, “Clustering via concave minimization,” in: M. C. Mozer, M. I. Jordan, and T. Petsche (eds.), Adv. in Neural Inform. Proces. Systems, MIT Press, Cambridge (1997), pp. 368–374.Google Scholar
  7. 7.
    L. Breiman, J. Friedman, R. Olsen, and C. Stone, Classification and Regression Tree, Wadsworth Inc., Belmont (1993).Google Scholar
  8. 8.
    E. G. Caiani, A. Porta, G. Baselli, et al., “Warped-average template technique to track on a cycle-by-cycle basis the cardiac filling phases on left ventricular volume,” IEEE Comput. in Cardiology, 25, No. 98, CH36292 (1998).Google Scholar
  9. 9.
    W. A. Chaovalitwongse, P. M. Pardalos, L. D. Iasemidis, et al., “Applications of global optimization and dynamical systems to prediction of epileptic seizures,” in: P.M. Pardalos, J.C. Sackellares, L. D. Iasemidis, and P. R. Carney (eds.), Quantitative Neuroscience, Kluwer, Dordrecht (2003), pp. 1–36.Google Scholar
  10. 10.
    W. A. Chaovalitwongse, P. M. Pardalos, and O. A. Prokoyev, “Reduction of multi-quadratic 0-1 programming problems to linear mixed 0-1 programming problems,” Oper. Res. Letters, 32, No. 6, 517–522 (2004).MATHCrossRefGoogle Scholar
  11. 11.
    W. A. Chaovalitwongse, P. M. Pardalos, and O. A. Prokoyev, “Electroencephalogram (EEG) time series classification: Applications in epilepsy,” Ann. Oper. Res., 148, No. 1, 227–250 (2006).MATHCrossRefGoogle Scholar
  12. 12.
    D. Dasgupta and S. Forrest, “Novelty detection in time series data using ideas from immunology,” Intern. Conf. on Intell. Systems (1999).Google Scholar
  13. 13.
    J. J. R. Diez and C. A. Gonzalez, “Applying boosting to similarity literals for time series classification,” in: Intern. Workshop on Multiple Classifier Systems (2000), pp. 210–219.Google Scholar
  14. 14.
    C. E. Elger and K. Lehnertz, “Seizure prediction by non-linear time series analysis of brain electrical activity,” Eur. J. Neurosci., 10, 786–789 (1998).CrossRefGoogle Scholar
  15. 15.
    G. M. Fung and O. L. Mangasarian, “Proximal support vector machines,” in: 7th ACM SIGKDD Intern. Conf. on Knowledge Discovery and Data Mining (2001).Google Scholar
  16. 16.
    D. M. Gavrila and L. S. Davis, “Towards 3-d model-based tracking and recognition of human movement: A multi-view approach,” in: Proc. Intern. Workshop on Autom. Face-and Gesture-Recognition (1995).Google Scholar
  17. 17.
    P. Geurts, “Pattern extraction for time series classification,” in: 5th Eur. Conf. on Principles of Data Mining and Knowledge Discovery (2001), pp. 115–127.Google Scholar
  18. 18.
    R. L. Grossman, C. Kamath, P. Kegelmeyer, et al., Data Mining for Scientific and Engineering Applications, Kluwer, Dordrecht (2001).MATHGoogle Scholar
  19. 19.
    D. J. Hand, H. Mannila, and P. Smyth, Principle of Data Mining, Bradford Books, Concord (2001).Google Scholar
  20. 20.
    C.-W. Hsu and C.-J. Lin, “A comparison of methods multi-class support vector machines,” IEEE Trans. on Neural Networks, 13, 415–425 (2002).CrossRefGoogle Scholar
  21. 21.
    A. S. Hurn, K. A. Lindsay, and C. A. Michie, “Modelling the lifespan of human t-lymphocite subsets,” Math. Biosciences, 143, 91–102 (1997).MATHCrossRefGoogle Scholar
  22. 22.
    F. J. Iannatilli and P. A. Rubin, “Feature selection for multiclass discrimination via mixed-integer linear programming,” IEEE Trans. on Pattern Analysis and Machine Learning, 25, 779–783 (2003).CrossRefGoogle Scholar
  23. 23.
    L. D. Iasemidis, “On the dynamics of the human brain in temporal lobe epilepsy,” PhD Thesis, Univ. of Michigan, Ann Arbor (1991).Google Scholar
  24. 24.
    L. D. Iasemidis, P. M. Pardalos, J. C. Sackellares, and D.-S. Shiau, “Quadratic binary programming and dynamical system approach to determine the predictability of epileptic seizures,” J. Comb. Optimiz., 5, 9–26 (2001).MATHCrossRefMathSciNetGoogle Scholar
  25. 25.
    L. D. Iasemidis, D.-S. Shiau, W. A. Chaovalitwongse, et al., “Adaptive epileptic seizure prediction system,” IEEE Trans. Biomed. Eng., 5, No. 5, 616–627 (2003).CrossRefGoogle Scholar
  26. 26.
    L. D. Iasemidis, H. P. Zaveri, J. C. Sackellares, and W. J. Williams, “Phase space analysis of eeg in temporal lobe epilepsy,” in: 10th Ann. Intern. Conf. “IEEE Eng. in Medicine and Biology Soc.” (1988), pp. 1201–1203.Google Scholar
  27. 27.
    E. Keogh, K. Chakrabarti, M. Pazzani, and S. Mehrotra, “Dimensionality reduction for fast similarity search in large time series databases,” Knowledge and Inform. Systems, 3, No. 3, 263–286 (2000).CrossRefGoogle Scholar
  28. 28.
    E. Keogh and S. Kasetty, “On the need for time series data mining benchmarks: A survey and empirical demonstration,” in: 8th ACM SIGKDD Intern. Conf. on Knowledge Discovery and Data Mining (2002), pp. 102–111.Google Scholar
  29. 29.
    E. Keogh and M. Pazzani, “An enhanced representation of time series which allows fast and accurate classification, clustering and relevance feedback,” in: 4th Intern. Conf. on Knowledge Discovery and Data Mining (1998), pp. 239–241.Google Scholar
  30. 30.
    E. Keogh and M. Pazzani, “Scaling up dynamic time warping for datamining applications,” in: Proc. 6th ACM SIGKDD Intern. Conf. on Knowledge Discovery and Data Mining (2000), pp. 285–289.Google Scholar
  31. 31.
    U. Krebel, “Pairwise classification and support vector machines,” in: Adv. in Kernel Methods-Support Vector Learning, MIT Press, Cambridge (1999), pp. 255–268.Google Scholar
  32. 32.
    K. Lehnertz and C. E. Elger, “Can epileptic seizures be predicted? Evidence from nonlinear time series analysis of brain electrical activity,” Phys. Rev. Lett., 80, 5019–5022 (1998).CrossRefGoogle Scholar
  33. 33.
    B. Litt, R. Esteller, J. Echauz, et al., “Epileptic seizures may begin hours in advance of clinical onset: A report of five patients,” Neuron, 30, 51–64 (2001).CrossRefGoogle Scholar
  34. 34.
    O. L. Mangasarian, “Linear and nonlinear separation of pattern by linear programming,” Oper. Res., 31, 445–453 (1965).MathSciNetGoogle Scholar
  35. 35.
    O. L. Mangasarian, W. N. Street, and W. H. Wolberg, “Breast cancer diagnosis and prognosis via linear programming,” Oper. Res., 43, No. 4, 570–577 (1995).MATHMathSciNetGoogle Scholar
  36. 36.
    J. Martinerie, C. Van Adam, and M. L. Van Quyen, “Epileptic seizures can be anticipated by non-linear analysis,” Nature Medicine, 4, 1173–1176 (1998).CrossRefGoogle Scholar
  37. 37.
    N. H. Packard, J. P. Crutchfield, and J. D. Farmer, “Geometry from time series,” Phys. Rev. Lett., 45, 712–716 (1980).CrossRefGoogle Scholar
  38. 38.
    P. M. Pardalos, W. A. Chaovalitwongse, L. D. Iasemidis, et al., “Seizure warning algorithm based on spatiotemporal dynamics of intracranial EEG,” Math. Program., 101, No. 2, 365–385 (2004).MATHCrossRefMathSciNetGoogle Scholar
  39. 39.
    J. R. Quinlan, C4.5: Programs for Machine Learning, Morgan Kaufmann, Orlando (1993).Google Scholar
  40. 40.
    M. L. Van Quyen, J. Martinerie, M. Baulac, and F. Varela, “Anticipating epileptic seizures in real time by non-linear analysis of similarity between eeg recordings,” Neuro Rep., 10, 2149–2155 (1999).Google Scholar
  41. 41.
    L. Rabiner and B. Juang, Fundamentals of Speech Recognition, Prentice Hall, Upper Saddle River (1993).Google Scholar
  42. 42.
    P. E. Rapp, I. D. Zimmerman, and A. M. Albano, “Experimental studies of chaotic neural behavior: Cellular activity and electroencephalographic signals,” in: H. G. Othmer (ed.), Nonlinear Oscillations in Biology and Chemistry, Springer, New York (1986), pp. 175–205.Google Scholar
  43. 43.
    S. Rüping, “SVM kernels for time series analysis,” in: R. Klinkenberg, S. Rüping, A. Fick, N. Henze, C. Herzog, R. Molitor, and O. Schröder (ed.), LLWA 01-Tagungsband der GI-Workshop-Woche Lernen-Lehren-Wissen-Adaptivitet (2001), pp. 43–50.Google Scholar
  44. 44.
    M. Schmill, T. Oates, and P. Cohen, “Learned models for continuous planning,” in: Proc. 7th Intern. Workshop on Artif. Intell. and Statist. (1999), pp. 278–282.Google Scholar
  45. 45.
    B. Scholkopf, C. Burges, and V. Vapnik, “Extracting support data for a given task,” in: Proc. 1st Intern. Conf. on Knowledge Discovery and Data Mining, AAAI Press, Menlo Park (1995).Google Scholar
  46. 46.
    B. Schölkopf, “The kernel trick for distances,” in: Techn. Rep., Microsoft Research (2000).Google Scholar
  47. 47.
    H. Shimodaira, K. Noma, M. Naka, and S. Sagayama, “Support vector machine with dynamic time-alignment kernel for speech recognition,” in: Proc. of Eurospeech (2001), pp. 1841–1844.Google Scholar
  48. 48.
    F. Takens, “Detecting strange attractors in turbulence,” D. A. Rand and L. S. Young (eds.), Dynamical Systems and Turbulence, Lecture Notes in Mathematics, Springer-Verlag, Berlin (1981).Google Scholar
  49. 49.
    V. N. Vapnik, The Nature of Statistical Learning, Springer, Berlin (1995).MATHGoogle Scholar
  50. 50.
    V. Wan and J. Carmichael, “Polynomial dynamic time warping kernel support vector machines for dysarthric speech recognition with sparse training data,” in: Proc. of Interspeech (2005), pp. 3321–3324.Google Scholar
  51. 51.
    A. Wolf, J. B. Swift, H. L. Swinney, and J. A. Vastano, “Determining Lyapunov exponents from a time series,” Physica D, 16, 285–317 (1985).MATHCrossRefMathSciNetGoogle Scholar
  52. 52.
    K. Yang and C. Shahabi, “A pca-based kernel for kernel pca on multivariate time series,” in: Proc. of ICDM 2005 Workshop on Temporal Data Mining: Algorithms, Theory and Applications held in conjunction with the 5th IEEE Intern. Conf. on Data Mining (ICDM’05) (2005).Google Scholar

Copyright information

© Springer Science+Business Media, Inc. 2008

Authors and Affiliations

  1. 1.Rutgers UniversityPiscatawayUSA
  2. 2.University of FloridaGainesvilleUSA

Personalised recommendations