Abstract Efficient online detection of similar patterns under arbitrary time scaling is a challenging problem in time series data mining. A modelmatching based segmental semiMarkov model is improved by introducing offset distribution, amplitude difference distribution and prepattern state. It overcomes the parameter estimation difficulty and the lack of robustness. The experimental results demonstrate that the advanced segmental semiMarkov model could rapidly and precisely detect scaling similar patterns under arbitrary time.
[1] Faloutsos C, Ranganathan M, Manolopoulos Y. Fast Subsequence Matching in TimeSeries Databases // Proc of the ACM SIGMOD International Conference on Management of Data. Minneapolis, USA, 1994: 419429 [2] Keogh E J, Pazzani M J. A Simple Dimensionality Reduction Technique for Fast Similarity Search in Large Time Series Databases // Proc of the 4th PacificAsia Conference on Knowledge Discover and Data Mining. Kyoto, Japan, 2000: 122133 [3] Ge X P, Smyth P. Deformable Markov Model Templates for TimeSeries Pattern Matching // Proc of the 6th ACM SIGKDD International Conference on Knowledge Discover and Data Mining. Boston, USA, 2000: 8190 [4] Meek C, Birmingham W P. The Dangers of Parsimony in QuerybyHumming Applications // Proc of the 4th International Symposium on Music Information Retrieval. Baltimore, USA, 2003: 5156 [5] Keogh E, Palpanas T, Zordan V B, et al. Indexing Large HumanMotion Databases // Proc of the 30th International Conference on Very Large Data Bases. Toronto, Canada, 2004: 780791 [6] Mandelbrot B B. Fractals and Scaling in Finance: Discontinuity, Concentration, Risk. New York, USA: SpringerVerlag, 2000 [7] Fu A W, Keogh E, Lau L Y H, et al. Scaling and Time Warping in Time Series Querying // Proc of the 31st International Conference on Very Large Data Bases. Trondheim, Norway, 2005: 649660 [8] Argyros T, Ermopoulos C. Efficient Subsequence Matching in Time Series Databases under Time and Amplitude Transformations // Proc of the 3rd IEEE International Conference on Data Mining. Melbourne, USA, 2003: 481484 [9] Rabiner L R, Rosenberg A E, Levinson S E. Considerations in Dynamic Time Warping Algorithms for Discrete Word Recognition. IEEE Trans on Acoustics, Speech and Signal Processing, 1978, 26(6): 575582 [10] Baum L E, Petrie T. Statistical Inference for Probabilistic Functions of Finite State Markov Chains. Annual of Mathematical Statistics, 1966, 37(6): 15541563 [11] Rabiner L R. A Tutorial on Hidden Markov Models and Selected Applications in Speech Recognition. Proc of the IEEE, 1989, 77(2): 257286 [12] Ferguson J D. Variable Duration Models for Speech // Proc of the Symposium on the Application of Hidden Markov Models to Text and Speech. Princeton, USA, 1980: 143179 [13] Ostendorf M, Digalakis V V, Kimball O A. From HMM's to Segmental Model: a Unified View of Stochastic Modeling for Speech Recognition. IEEE Trans on Speech and Audio Processing, 1996, 4(5): 360378 [14] Ge X P, Smyth P. Hidden Markov Models for Endpoint Detection in Plasma Etch Processes. Technical Report, UCIICS 0154, Irvine, USA: University of California. Department of Information and Computer Science, 2001 [15] Jia Sen, Qian Yuntao, Dai Guang. An Advance Segmental SemiMarkov Model Based Online Series Pattern Detection // Proc of the 17th International Conference on Pattern Recognition. Cambridge, UK, 2004, Ⅲ: 634 637
2007, Vol. 20 
