The Sensitivity of the Modified Viterbi Algorithm to the Source Statistics

Rajjan Shinghal, Godfried Toussaint

Research output: Contribution to journalArticle

Abstract

The modified Viterbi algorithm is a powerful, and increasingly used, tool for using contextual information in text recognition in its various forms. As yet, no known studies have been published concerning its robustness with respect to source statistics. This paper describes experiments performed to determine the sensitivity of the algorithm to variations in source statistics. The results of the experiments show that a character-recognition machine incorporating the modified Viterbi algorithm, using N-gram statistics estimated from source A does not deteriorate in performance when operating on a passage from source B even if A and B differ significantly in TV-gram distributions or entropy.

Original languageEnglish (US)
Pages (from-to)181-185
Number of pages5
JournalIEEE Transactions on Pattern Analysis and Machine Intelligence
VolumePAMI-2
Issue number2
DOIs
StatePublished - Jan 1 1980

Fingerprint

Viterbi Algorithm
Viterbi algorithm
Statistics
Character recognition
N-gram
Character Recognition
Entropy
Experiments
Experiment
Robustness

Keywords

  • Character recognition
  • contextural information
  • entropy
  • Markov process
  • natural language statistics
  • robustness tests
  • text processing
  • Viterbi algorithm

ASJC Scopus subject areas

  • Software
  • Computer Vision and Pattern Recognition
  • Computational Theory and Mathematics
  • Artificial Intelligence
  • Applied Mathematics

Cite this

The Sensitivity of the Modified Viterbi Algorithm to the Source Statistics. / Shinghal, Rajjan; Toussaint, Godfried.

In: IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. PAMI-2, No. 2, 01.01.1980, p. 181-185.

Research output: Contribution to journalArticle

@article{29f109c3dcfb44b8a687e8fbabe7b7e1,
title = "The Sensitivity of the Modified Viterbi Algorithm to the Source Statistics",
abstract = "The modified Viterbi algorithm is a powerful, and increasingly used, tool for using contextual information in text recognition in its various forms. As yet, no known studies have been published concerning its robustness with respect to source statistics. This paper describes experiments performed to determine the sensitivity of the algorithm to variations in source statistics. The results of the experiments show that a character-recognition machine incorporating the modified Viterbi algorithm, using N-gram statistics estimated from source A does not deteriorate in performance when operating on a passage from source B even if A and B differ significantly in TV-gram distributions or entropy.",
keywords = "Character recognition, contextural information, entropy, Markov process, natural language statistics, robustness tests, text processing, Viterbi algorithm",
author = "Rajjan Shinghal and Godfried Toussaint",
year = "1980",
month = "1",
day = "1",
doi = "10.1109/TPAMI.1980.4766998",
language = "English (US)",
volume = "PAMI-2",
pages = "181--185",
journal = "IEEE Transactions on Pattern Analysis and Machine Intelligence",
issn = "0162-8828",
publisher = "IEEE Computer Society",
number = "2",

}

TY - JOUR

T1 - The Sensitivity of the Modified Viterbi Algorithm to the Source Statistics

AU - Shinghal, Rajjan

AU - Toussaint, Godfried

PY - 1980/1/1

Y1 - 1980/1/1

N2 - The modified Viterbi algorithm is a powerful, and increasingly used, tool for using contextual information in text recognition in its various forms. As yet, no known studies have been published concerning its robustness with respect to source statistics. This paper describes experiments performed to determine the sensitivity of the algorithm to variations in source statistics. The results of the experiments show that a character-recognition machine incorporating the modified Viterbi algorithm, using N-gram statistics estimated from source A does not deteriorate in performance when operating on a passage from source B even if A and B differ significantly in TV-gram distributions or entropy.

AB - The modified Viterbi algorithm is a powerful, and increasingly used, tool for using contextual information in text recognition in its various forms. As yet, no known studies have been published concerning its robustness with respect to source statistics. This paper describes experiments performed to determine the sensitivity of the algorithm to variations in source statistics. The results of the experiments show that a character-recognition machine incorporating the modified Viterbi algorithm, using N-gram statistics estimated from source A does not deteriorate in performance when operating on a passage from source B even if A and B differ significantly in TV-gram distributions or entropy.

KW - Character recognition

KW - contextural information

KW - entropy

KW - Markov process

KW - natural language statistics

KW - robustness tests

KW - text processing

KW - Viterbi algorithm

UR - http://www.scopus.com/inward/record.url?scp=0018995386&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0018995386&partnerID=8YFLogxK

U2 - 10.1109/TPAMI.1980.4766998

DO - 10.1109/TPAMI.1980.4766998

M3 - Article

AN - SCOPUS:0018995386

VL - PAMI-2

SP - 181

EP - 185

JO - IEEE Transactions on Pattern Analysis and Machine Intelligence

JF - IEEE Transactions on Pattern Analysis and Machine Intelligence

SN - 0162-8828

IS - 2

ER -