Abstract

A method, implemented on a computer having a fixed amount of memory and CPU resources, for analyzing a sequence of data units derived from a dynamic system to which new data units may be added by classifying the data units, is disclosed. The method comprises determining the similarity of the data units being part of the sequence of data units by calculating the distance between all pairs of data units in a data space. The method further comprises classifying the data units by assigning labels to the data units such that, if the distance of a data unit which is to be classified to any other data unit exceeds a threshold, a new label is assigned to the data unit to be classified. Also, if the threshold is not exceeded, the label of the data unit being closest to the data unit to be classified is assigned to the data unit to be classified.

Other References

• Wilcox et al., “Segmentation of Speech Using Speaker Identificaiton”, Jan. 1994, IEEE, vol: 78, No. 3, pp. 161-164.
• Kristjansson et al., “Towards Non-Stationary Model-Based Noise Adaptation For Large Vocabulary Speech Recognition”, University of Waterloo, CA, Jan. 2000, pp. 1-4.
• Schott, “Implementation of a Two-State Viterbi Decoder with Embedded Decision Feedback”, Jan. 1992, IEEE, pp. 181-184.
• Bengio, et. al, “An Input Output HMM Architecture.”
• Bishop, “Neural Networks for Pattern Recognition, section 2.5.3,” (Apr. 18, 1995).
• Cacciatore, et al, “Mixtures of Controllers for Jump Linear and Non-Linear Plants.”
• Cominetti, et al., “Stable Exponential-Penalty Algorithm with Superlinear Convergence,” JOTA , Plenum Publishing Corporation, vol: 83 (No. 2), p. 285-309, (Apr. 18, 1994).
• Fancourt, et al., “A Neighborhood Map of Competing One Step Predictors for Piecewise Segmentation and Identification of Time Series.”
• Husmeier, “Learning non-stationary conditional probability distributions,” Neural Networks, Pergamon, p. 287-290, (Apr. 18, 2000).
• Jacobs, et al., “Adaptive Mixtures of Local Experts,” Neural Computation, Mass. Inst. of Tech. (USA), p. 79-87, (Apr. 18, 1997).
• Kehagias, et al, “Time series Segmentation using Predictive Modular Neural Networks,” Neural Computation,Mass. Inst. of Tech. (USA), p. 1691-1710, (Apr. 18, 1997).
• Kohlmorgen, et al., “Improving short term prediction with competing experts.”
• Kohlmorgen, et al, “Segmentation and Identification of Drifting Dynamical Systems.”
• Kohlmorgen, et al, “Identification of nonstationary dynamics in physiological recordings,” Biological Cybernetics, Springer Verlag, p. 73-84, (Apr. 18, 2000).
• Liehr, et al., “Hidden Markov Mixtures of Experts with an Application to EEG Recordings from Sleep.”
• Mackey, et al., “Oscillation and Chaos in Physiological Control Systems,” Science, p. 287-289, (Jul. 15, 1977).
• Pawelzik, et al, “Annealed Competition of Experts for a Segmentation and Classification of Switching Dynamics,” Neural Computation, Mass. Inst. of Tech. (USA), p. 340-356, (Apr. 18, 1996).
• Ramamurti, et al., “Structurally Adaptive Modular Networks for Nonstationary Environments,” Trans. on Neural Networks, IEEE (USA), vol. 10 (No. 1), p. 152-160, (Apr. 18, 1999.
• Shi, et al., “Taking Time Seriously. Hidden Markov Experts Applied to Financial Engineering,” CIFEr, IEEE (USA), (Mar. 30, 1997).
• Takena, Floria, “Detecting strange attractors in turbulence.”
• Packard, et al., “Geometry from a Time Series,” Physical Review Letters, The American Physical Society (USA), vol. 45 (No. 9), p. 712-716, (Apr. 18, 1980).
• MacQueen, “Some Methods for Classification and Analysis of Multivariate Observations,” Fifth Berkeley Symposium, p. 281-297.
• Theorodridis, S. et al., “Pattern Recognition (XP002244947),” Academic Press (San Diego, CA), p. 372-373, (Aug. 14, 1999).
• Svendsen, T. et al, “Automatic alignment of phonemic labels in continuous speech,” COST 249, p. 1-6, (Mar. 7, 1995).
• Estivill-Castro, et al, “Hybrid Optimization for Clustering in Data Mining,” Proceeding of X Claio, p. 1-9, (Aug. 14, 2000).