JOURNAL BROWSE
Search
Advanced SearchSearch Tips
Discrete HMM Training Algorithm for Incomplete Time Series Data
facebook(new window)  Pirnt(new window) E-mail(new window) Excel Download
 Title & Authors
Discrete HMM Training Algorithm for Incomplete Time Series Data
Sin, Bong-Kee;
  PDF(new window)
 Abstract
Hidden Markov Model is one of the most successful and popular tools for modeling real world sequential data. Real world signals come in a variety of shapes and variabilities, among which temporal and spectral ones are the prime targets that the HMM aims at. A new problem that is gaining increasing attention is characterizing missing observations in incomplete data sequences. They are incomplete in that there are holes or omitted measurements. The standard HMM algorithms have been developed for complete data with a measurements at each regular point in time. This paper presents a modified algorithm for a discrete HMM that allows substantial amount of omissions in the input sequence. Basically it is a variant of Baum-Welch which explicitly considers the case of isolated or a number of omissions in succession. The algorithm has been tested on online handwriting samples expressed in direction codes. An extensive set of experiments show that the HMM so modeled are highly flexible showing a consistent and robust performance regardless of the amount of omissions.
 Keywords
Hidden Markov Model;Training Algorithm;Discrete HMM;Incomplete Sequence Data;
 Language
Korean
 Cited by
 References
1.
L. Rabiner, "A Tutorial on Hidden Markov Models and Selected Applications in Speech Recognition," Proceedings of the IEEE, Vol. 77, No. 2, pp. 257-286, 1989. crossref(new window)

2.
B.-K. Sin, "Recognizing Hand Digit Gestures Using Stochastic Models," Journal of Korea Multimedia Society, Vol. 11, No. 6, pp. 807-815, 2008. 6.

3.
B.-K. Sin and K.-R. Kwon, "Feature Space Analysis of Human Gait Dynamics in Single View Video," Journal of Korea Multimedia Society, Vol. 13, No. 12, pp. 1778-1785, 2010. 12.

4.
F. Torre, D. Pitchford, P. Brown, and L. Terveen, "Matching GPS Traces to (Possibly) Incomplete Map Data: Bridging Map Building and Map Matching," Proceeding of ACM SIGSPATIAL GIS'12, pp. 546-549, 2012.

5.
R. Little and D.B. Rubin, Statistical Analysis with Missing Data, John Wiley & Sons, Inc., NY, USA, p. 408, 1987.

6.
M. Cooke, P. Green, and M. Crawford, "Handling Missing Data in Speech Recognition," Proceeding of International Conference on Spoken Language Processing, pp. 1555-1558, 1994.

7.
A.C. Morris, M.P. Cooke, and P.D. Green, "Some Solution to the Missing Feature Problem in Data Classification, with Application to Noise Robust ASR," Proceeding of International Conference on Acoustics, Speech and Signal Processing, Vol. 2, pp. 737-740, 1998.

8.
S. Parveen and P.D. Green, "Speech Recognition with Missing Data using Recurrent Neural Network," in Advances in Neural Information Processing Systems, 14, pp. 1189-1195, 2001.

9.
L.E. Baum and J.A. Eagon, "An Inequality with Applications to Statistical Estimation for Probabilistic Functions of Markov Processes and to a Model for Ecology," Bulletin of the American Mathematical Society, Vol. 73, No. 3, pp. 360-363, 1967. crossref(new window)

10.
A.P. Dempster, N.M. Rubin, and D.B. Rubin, "Maximum Likelihood from Incomplete Data via the EM Algorithm," Journal of the Royal Statistical Society, Series B, Vol. 39, No. 1, pp. 1-38, 1977.