Language identification using wavelet transform and artificial neural network

Shawki Al-Dubaee; Nesar  Ahmad; Jan  Martinovic̆; Vaclav  Snás̆el

doi:10.1109/CASoN.2010.121

Language identification using wavelet transform and artificial neural network

Shawki Al-Dubaee, Nesar Ahmad, Jan Martinovic̆, Vaclav Snás̆el

National Manufacturing Institute Scotland

Research output: Contribution to conference › Paper › peer-review

4 Citations (Scopus)

Abstract

In traditional language identification methods, it is not so easy for search engines to find relevant language database of a given query. Therefore, there is a need to identify the relevant user's natural language query of unknown document database in a better way by automatic language identification. This novel approach presents an automatic method for classification of English and Arabic language identification. The classifier used is a three-layered feed-forward artificial neural network and the feature vector is formed by calculating the wavelet coefficients. Three wavelet decomposition functions (filters), namely Haar, Bior 2.2 and Bior 3.1 have been used to extract the feature vector set and their performance has been compared.

Original language	English
Number of pages	6
DOIs	https://doi.org/10.1109/CASoN.2010.121
Publication status	Published - 28 Sept 2010
Event	2010 International Conference on Computational Aspects of Social Networks - Taiyuan, China Duration: 26 Sept 2010 → 28 Sept 2010

Conference

Conference	2010 International Conference on Computational Aspects of Social Networks
Country/Territory	China
City	Taiyuan
Period	26/09/10 → 28/09/10

Keywords

wavelet transform
artificial neural network
language identification
cross language

Access to Document

10.1109/CASoN.2010.121

Cite this

@conference{7cfb2b9485184a3d90a3405317cea3ba,

title = "Language identification using wavelet transform and artificial neural network",

abstract = "In traditional language identification methods, it is not so easy for search engines to find relevant language database of a given query. Therefore, there is a need to identify the relevant user's natural language query of unknown document database in a better way by automatic language identification. This novel approach presents an automatic method for classification of English and Arabic language identification. The classifier used is a three-layered feed-forward artificial neural network and the feature vector is formed by calculating the wavelet coefficients. Three wavelet decomposition functions (filters), namely Haar, Bior 2.2 and Bior 3.1 have been used to extract the feature vector set and their performance has been compared.",

keywords = "wavelet transform, artificial neural network, language identification, cross language",

author = "Shawki Al-Dubaee and Nesar Ahmad and Jan Martinovi{\u c} and Vaclav Sn{\'a}{\u s}el",

note = "(Nominated to AWARD); 2010 International Conference on Computational Aspects of Social Networks ; Conference date: 26-09-2010 Through 28-09-2010",

year = "2010",

month = sep,

day = "28",

doi = "10.1109/CASoN.2010.121",

language = "English",

}

TY - CONF

T1 - Language identification using wavelet transform and artificial neural network

AU - Al-Dubaee, Shawki

AU - Ahmad, Nesar

AU - Martinovic̆, Jan

AU - Snás̆el, Vaclav

N1 - (Nominated to AWARD)

PY - 2010/9/28

Y1 - 2010/9/28

N2 - In traditional language identification methods, it is not so easy for search engines to find relevant language database of a given query. Therefore, there is a need to identify the relevant user's natural language query of unknown document database in a better way by automatic language identification. This novel approach presents an automatic method for classification of English and Arabic language identification. The classifier used is a three-layered feed-forward artificial neural network and the feature vector is formed by calculating the wavelet coefficients. Three wavelet decomposition functions (filters), namely Haar, Bior 2.2 and Bior 3.1 have been used to extract the feature vector set and their performance has been compared.

AB - In traditional language identification methods, it is not so easy for search engines to find relevant language database of a given query. Therefore, there is a need to identify the relevant user's natural language query of unknown document database in a better way by automatic language identification. This novel approach presents an automatic method for classification of English and Arabic language identification. The classifier used is a three-layered feed-forward artificial neural network and the feature vector is formed by calculating the wavelet coefficients. Three wavelet decomposition functions (filters), namely Haar, Bior 2.2 and Bior 3.1 have been used to extract the feature vector set and their performance has been compared.

KW - wavelet transform

KW - artificial neural network

KW - language identification

KW - cross language

U2 - 10.1109/CASoN.2010.121

DO - 10.1109/CASoN.2010.121

M3 - Paper

T2 - 2010 International Conference on Computational Aspects of Social Networks

Y2 - 26 September 2010 through 28 September 2010

ER -

Language identification using wavelet transform and artificial neural network

Abstract

Conference

Keywords

Access to Document

Fingerprint

Cite this