Adaptive pattern classification for symbolic dynamic systems

Yicheng Wen, Kushal Mukherjee, Asok Ray

Research output: Contribution to journalArticle

18 Citations (Scopus)

Abstract

This paper addresses pattern classification in dynamical systems, where the underlying algorithms are formulated in the symbolic domain and the patterns are constructed from symbol strings as probabilistic finite state automata (PFSA) with (possibly) diverse algebraic structures. A combination of Dirichlet and multinomial distributions is used to model the uncertainties due to the (finite-length) string approximation of symbol sequences in both training and testing phases of pattern classification. The classifier algorithm follows the structure of a Bayes model and has been validated on a simulation test bed. The results of numerical simulation are presented for several examples.

Original languageEnglish (US)
Pages (from-to)252-260
Number of pages9
JournalSignal Processing
Volume93
Issue number1
DOIs
StatePublished - Jan 1 2013

Fingerprint

Pattern recognition
Dynamical systems
Finite automata
Classifiers
Computer simulation
Testing
Uncertainty

All Science Journal Classification (ASJC) codes

  • Control and Systems Engineering
  • Software
  • Signal Processing
  • Computer Vision and Pattern Recognition
  • Electrical and Electronic Engineering

Cite this

Wen, Yicheng ; Mukherjee, Kushal ; Ray, Asok. / Adaptive pattern classification for symbolic dynamic systems. In: Signal Processing. 2013 ; Vol. 93, No. 1. pp. 252-260.
@article{f244596e602e432f9cedc44a3340084e,
title = "Adaptive pattern classification for symbolic dynamic systems",
abstract = "This paper addresses pattern classification in dynamical systems, where the underlying algorithms are formulated in the symbolic domain and the patterns are constructed from symbol strings as probabilistic finite state automata (PFSA) with (possibly) diverse algebraic structures. A combination of Dirichlet and multinomial distributions is used to model the uncertainties due to the (finite-length) string approximation of symbol sequences in both training and testing phases of pattern classification. The classifier algorithm follows the structure of a Bayes model and has been validated on a simulation test bed. The results of numerical simulation are presented for several examples.",
author = "Yicheng Wen and Kushal Mukherjee and Asok Ray",
year = "2013",
month = "1",
day = "1",
doi = "10.1016/j.sigpro.2012.08.002",
language = "English (US)",
volume = "93",
pages = "252--260",
journal = "Signal Processing",
issn = "0165-1684",
publisher = "Elsevier",
number = "1",

}

Adaptive pattern classification for symbolic dynamic systems. / Wen, Yicheng; Mukherjee, Kushal; Ray, Asok.

In: Signal Processing, Vol. 93, No. 1, 01.01.2013, p. 252-260.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Adaptive pattern classification for symbolic dynamic systems

AU - Wen, Yicheng

AU - Mukherjee, Kushal

AU - Ray, Asok

PY - 2013/1/1

Y1 - 2013/1/1

N2 - This paper addresses pattern classification in dynamical systems, where the underlying algorithms are formulated in the symbolic domain and the patterns are constructed from symbol strings as probabilistic finite state automata (PFSA) with (possibly) diverse algebraic structures. A combination of Dirichlet and multinomial distributions is used to model the uncertainties due to the (finite-length) string approximation of symbol sequences in both training and testing phases of pattern classification. The classifier algorithm follows the structure of a Bayes model and has been validated on a simulation test bed. The results of numerical simulation are presented for several examples.

AB - This paper addresses pattern classification in dynamical systems, where the underlying algorithms are formulated in the symbolic domain and the patterns are constructed from symbol strings as probabilistic finite state automata (PFSA) with (possibly) diverse algebraic structures. A combination of Dirichlet and multinomial distributions is used to model the uncertainties due to the (finite-length) string approximation of symbol sequences in both training and testing phases of pattern classification. The classifier algorithm follows the structure of a Bayes model and has been validated on a simulation test bed. The results of numerical simulation are presented for several examples.

UR - http://www.scopus.com/inward/record.url?scp=84866431880&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84866431880&partnerID=8YFLogxK

U2 - 10.1016/j.sigpro.2012.08.002

DO - 10.1016/j.sigpro.2012.08.002

M3 - Article

AN - SCOPUS:84866431880

VL - 93

SP - 252

EP - 260

JO - Signal Processing

JF - Signal Processing

SN - 0165-1684

IS - 1

ER -