Abstract
This paper describes a form of Markov chain, called a constrained Markov network, and its inference from finite-length sequences over a finite alphabet as a structural/statistical model of a class of strings for purposes of pattern analysis and recognition. In particular, we describe how the inference can be based on string alignments computed optimally by dynamic programming using an integer frequency-count disagreement cost function. We also discuss systematic reduction of network size by “pruning away” stages associated with low probability of observable symbols. Empirical results are reported for sequences representing band patterns in human chromosomes.
Supported in part by a Professional Development Award from the University of Tennessee.
This is a preview of subscription content, log in via an institution.
Preview
Unable to display preview. Download preview PDF.
References
L. Miclet, “Grammatical inference,” in Syntactic and Structural Pattern Recognition Theory and Applications (H. Bunke and A. Sanfeliu, eds.), pp. 237–290, World Scientific, 1990.
J. Gregor, “Data-driven inductive inference of finite-state automata,” Int. J. Pattern Recogn. Artif. Intell., vol. 8, pp. 305–322, 1994.
M. G. Thomason and J. Gregor, “Inference of Markov chain models of sequences,” Current Topics in Pattern Recognition Research, pp. 163–173, 1995.
M. G. Thomason and E. Granum, “Dynamic programming inference of Markov networks from finite sets of sample strings,” IEEE Trans. PAMI, vol. 8, pp. 491–501, 1986.
H. Rulot and E. Vidal, “Modelling (sub)string-length based constraints through a grammatical inference method,” in Pattern Recognition Theory and Applications (P. A. Devijver and J. Kittler, eds.), pp. 451–459, Springer-Verlag, 1987.
H. Rulot and E. Vidal, “An efficient algorithm for the inference of circuit-free automata,” in Syntactic and Structural Pattern Recognition (G. Ferratè, T. Pavlidis, A. Sanfeliu, and H. Bunke, eds.), pp. 173–184, Springer-Verlag, 1988.
C. E. Guthrie, J. Gregor, and M. G. Thomason, “Constrained Markov networks for automated analysis of G-banded chromosomes,” Comput. Biol. Medicine, vol. 23, pp. 105–114, 1993.
H. Bunke and D. Pasche, “Parsing multivalued strings and its application to image and waveform recognition,” in Structural Pattern Analysis (R. Mohr, T. Pavlidis, and A. Sanfeliu, eds.), pp. 1–17, World Scientific, 1989.
R. A. Wagner and M. J. Fischer, “The string-to-string correction problem,” J. Assoc. Comput. Mach., vol. 21, pp. 168–173, 1974.
D. Sankoff and J. B. Kruskal, eds., Time Warps, String Edits, and Macromolecules: The Theory and Practice of Sequence Comparisons. Addison-Wesley, 1983.
E. Granum, M. G. Thomason, and J. Gregor, “On the use of automatically inferred Markov networks for chromosome analysis,” in Automation of Cytogenetics (C. Lundsteen and J. Piper, eds.), pp. 233–251, Springer-Verlag, 1989.
E. Granum and M. G. Thomason, “Automatically inferred Markov network models for classification of chromosomal band pattern structures,” Cytometry, vol. 11, pp. 26–39, 1990.
J. Gregor and M. G. Thomason, “Hybrid pattern recognition using Markov networks,” IEEE Trans. PAMI, vol. 15, pp. 651–656, 1993.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1996 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Gregor, J., Thomason, M.G. (1996). A disagreement count scheme for inference of constrained Markov networks. In: Miclet, L., de la Higuera, C. (eds) Grammatical Interference: Learning Syntax from Sentences. ICGI 1996. Lecture Notes in Computer Science, vol 1147. Springer, Berlin, Heidelberg. https://doi.org/10.1007/BFb0033352
Download citation
DOI: https://doi.org/10.1007/BFb0033352
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-61778-5
Online ISBN: 978-3-540-70678-6
eBook Packages: Springer Book Archive