Neo-statistical methods of recognition: Neural network, hidden Markov model, and their hybridization

Sung Bae Cho

doi:10.1016/0920-5489(94)90013-2

Neo-statistical methods of recognition: Neural network, hidden Markov model, and their hybridization

Sung Bae Cho

College of Computing

Research output: Contribution to journal › Article › peer-review

1 Citation (Scopus)

Abstract

Recently, several new methodologies based on statistical schemes have appeared that produce successful results in the field of pattern recognition. This paper presents two of them, neural network classifiers and hidden Markov models (HMMs), and further proposes a hybrid of both. The hybrid method exploits the discrimination capability of neural network classifiers while using HMM formalism to capture the dynamics of input patterns. The presented methods are applied to the recognition of on-line handwritten characters. Experimental results with a large set of multi-writer data confirm the usefulness of more sophisticated methodologies such as multistage schemes and multiple network schemes as well as the hybrid method.

Original language	English
Pages (from-to)	221-229
Number of pages	9
Journal	Computer Standards and Interfaces
Volume	16
Issue number	3
DOIs	https://doi.org/10.1016/0920-5489(94)90013-2
Publication status	Published - 1994 Jul

Bibliographical note

Funding Information:
* Email: sbcho(~'hip.atr.co.jp, sbcho@gorai.kaist.ac.kr * This work was supported in part by a grant from the Korea Science and Engineering Foundation (KOSEF) and Center for Artificial Intelligence Research (CAIR), the Engineering Research Center (ERC) of Excellence Program.

All Science Journal Classification (ASJC) codes

Software
Hardware and Architecture
Law

Access to Document

10.1016/0920-5489(94)90013-2

Cite this

@article{538e79dd17854043bb5b3e99cf739866,

title = "Neo-statistical methods of recognition: Neural network, hidden Markov model, and their hybridization",

abstract = "Recently, several new methodologies based on statistical schemes have appeared that produce successful results in the field of pattern recognition. This paper presents two of them, neural network classifiers and hidden Markov models (HMMs), and further proposes a hybrid of both. The hybrid method exploits the discrimination capability of neural network classifiers while using HMM formalism to capture the dynamics of input patterns. The presented methods are applied to the recognition of on-line handwritten characters. Experimental results with a large set of multi-writer data confirm the usefulness of more sophisticated methodologies such as multistage schemes and multiple network schemes as well as the hybrid method.",

author = "Cho, {Sung Bae}",

note = "Funding Information: * Email: sbcho(~'hip.atr.co.jp, sbcho@gorai.kaist.ac.kr * This work was supported in part by a grant from the Korea Science and Engineering Foundation (KOSEF) and Center for Artificial Intelligence Research (CAIR), the Engineering Research Center (ERC) of Excellence Program.",

year = "1994",

month = jul,

doi = "10.1016/0920-5489(94)90013-2",

language = "English",

volume = "16",

pages = "221--229",

journal = "Computer Standards and Interfaces",

issn = "0920-5489",

publisher = "Elsevier",

number = "3",

}

TY - JOUR

T1 - Neo-statistical methods of recognition

T2 - Neural network, hidden Markov model, and their hybridization

AU - Cho, Sung Bae

N1 - Funding Information: * Email: sbcho(~'hip.atr.co.jp, sbcho@gorai.kaist.ac.kr * This work was supported in part by a grant from the Korea Science and Engineering Foundation (KOSEF) and Center for Artificial Intelligence Research (CAIR), the Engineering Research Center (ERC) of Excellence Program.

PY - 1994/7

Y1 - 1994/7

N2 - Recently, several new methodologies based on statistical schemes have appeared that produce successful results in the field of pattern recognition. This paper presents two of them, neural network classifiers and hidden Markov models (HMMs), and further proposes a hybrid of both. The hybrid method exploits the discrimination capability of neural network classifiers while using HMM formalism to capture the dynamics of input patterns. The presented methods are applied to the recognition of on-line handwritten characters. Experimental results with a large set of multi-writer data confirm the usefulness of more sophisticated methodologies such as multistage schemes and multiple network schemes as well as the hybrid method.

AB - Recently, several new methodologies based on statistical schemes have appeared that produce successful results in the field of pattern recognition. This paper presents two of them, neural network classifiers and hidden Markov models (HMMs), and further proposes a hybrid of both. The hybrid method exploits the discrimination capability of neural network classifiers while using HMM formalism to capture the dynamics of input patterns. The presented methods are applied to the recognition of on-line handwritten characters. Experimental results with a large set of multi-writer data confirm the usefulness of more sophisticated methodologies such as multistage schemes and multiple network schemes as well as the hybrid method.

UR - http://www.scopus.com/inward/record.url?scp=0028466851&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0028466851&partnerID=8YFLogxK

U2 - 10.1016/0920-5489(94)90013-2

DO - 10.1016/0920-5489(94)90013-2

M3 - Article

AN - SCOPUS:0028466851

SN - 0920-5489

VL - 16

SP - 221

EP - 229

JO - Computer Standards and Interfaces

JF - Computer Standards and Interfaces

IS - 3

ER -

Neo-statistical methods of recognition: Neural network, hidden Markov model, and their hybridization

Abstract

Bibliographical note

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this