ADAM: Analyzer for Dialectal Arabic Morphology

Wael Salloum; Nizar Habash

doi:10.1016/j.jksuci.2014.06.010

ADAM: Analyzer for Dialectal Arabic Morphology

Wael Salloum, Nizar Habash

Computer Science

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

Abstract

While Modern Standard Arabic (MSA) has many resources, Arabic Dialects, the primarily spoken local varieties of Arabic, are quite impoverished in this regard. In this article, we present ADAM (Analyzer for Dialectal Arabic Morphology). ADAM is a poor man's solution to quickly develop morphological analyzers for dialectal Arabic. ADAM has roughly half the out-of-vocabulary rate of a state-of-the-art MSA analyzer and is comparable in its recall performance to an Egyptian dialectal morphological analyzer that took years and expensive resources to build.

Original language	English (US)
Pages (from-to)	372-378
Number of pages	7
Journal	Journal of King Saud University - Computer and Information Sciences
Volume	26
Issue number	4
DOIs	https://doi.org/10.1016/j.jksuci.2014.06.010
State	Published - 2014

Keywords

Arabic morphology
Arabic natural language processing
Dialectal Arabic
Machine translation

ASJC Scopus subject areas

Computer Science(all)

Access to Document

10.1016/j.jksuci.2014.06.010

Cite this

@article{8edd230e931a4188ae690fcc35dac1b0,

title = "ADAM: Analyzer for Dialectal Arabic Morphology",

abstract = "While Modern Standard Arabic (MSA) has many resources, Arabic Dialects, the primarily spoken local varieties of Arabic, are quite impoverished in this regard. In this article, we present ADAM (Analyzer for Dialectal Arabic Morphology). ADAM is a poor man's solution to quickly develop morphological analyzers for dialectal Arabic. ADAM has roughly half the out-of-vocabulary rate of a state-of-the-art MSA analyzer and is comparable in its recall performance to an Egyptian dialectal morphological analyzer that took years and expensive resources to build.",

keywords = "Arabic morphology, Arabic natural language processing, Dialectal Arabic, Machine translation",

author = "Wael Salloum and Nizar Habash",

note = "Funding Information: This research was supported by the Defense Advanced Research Projects Agency (DARPA) GALE program, contract HR0011-06-C-0022 , and the DARPA BOLT program, contract No. HR0011-12-C-0014. Any opinions, findings, conclusions or recommendations expressed in this work are those of the authors and do not necessarily reflect the view of DARPA. We would like to thank John Makhoul and Spyros Matsoukas for helpful discussions and feedback and for providing us with the data we used for initial analysis of dialectal phenomena. Publisher Copyright: {\textcopyright} 2014 The Authors.",

year = "2014",

doi = "10.1016/j.jksuci.2014.06.010",

language = "English (US)",

volume = "26",

pages = "372--378",

journal = "Journal of King Saud University - Computer and Information Sciences",

issn = "1319-1578",

publisher = "King Saud University",

number = "4",

}

TY - JOUR

T1 - ADAM

T2 - Analyzer for Dialectal Arabic Morphology

AU - Salloum, Wael

AU - Habash, Nizar

N1 - Funding Information: This research was supported by the Defense Advanced Research Projects Agency (DARPA) GALE program, contract HR0011-06-C-0022 , and the DARPA BOLT program, contract No. HR0011-12-C-0014. Any opinions, findings, conclusions or recommendations expressed in this work are those of the authors and do not necessarily reflect the view of DARPA. We would like to thank John Makhoul and Spyros Matsoukas for helpful discussions and feedback and for providing us with the data we used for initial analysis of dialectal phenomena. Publisher Copyright: © 2014 The Authors.

PY - 2014

Y1 - 2014

N2 - While Modern Standard Arabic (MSA) has many resources, Arabic Dialects, the primarily spoken local varieties of Arabic, are quite impoverished in this regard. In this article, we present ADAM (Analyzer for Dialectal Arabic Morphology). ADAM is a poor man's solution to quickly develop morphological analyzers for dialectal Arabic. ADAM has roughly half the out-of-vocabulary rate of a state-of-the-art MSA analyzer and is comparable in its recall performance to an Egyptian dialectal morphological analyzer that took years and expensive resources to build.

AB - While Modern Standard Arabic (MSA) has many resources, Arabic Dialects, the primarily spoken local varieties of Arabic, are quite impoverished in this regard. In this article, we present ADAM (Analyzer for Dialectal Arabic Morphology). ADAM is a poor man's solution to quickly develop morphological analyzers for dialectal Arabic. ADAM has roughly half the out-of-vocabulary rate of a state-of-the-art MSA analyzer and is comparable in its recall performance to an Egyptian dialectal morphological analyzer that took years and expensive resources to build.

KW - Arabic morphology

KW - Arabic natural language processing

KW - Dialectal Arabic

KW - Machine translation

UR - http://www.scopus.com/inward/record.url?scp=85006210731&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85006210731&partnerID=8YFLogxK

U2 - 10.1016/j.jksuci.2014.06.010

DO - 10.1016/j.jksuci.2014.06.010

M3 - Article

AN - SCOPUS:85006210731

SN - 1319-1578

VL - 26

SP - 372

EP - 378

JO - Journal of King Saud University - Computer and Information Sciences

JF - Journal of King Saud University - Computer and Information Sciences

IS - 4

ER -

ADAM: Analyzer for Dialectal Arabic Morphology

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Cite this