• DocumentCode
    234368
  • Title

    An XML database for modern standard Arabic (MSA) verbs generated from triliteral roots

  • Author

    Tahir, Youssef

  • Author_Institution
    Ecole Nat. Super. d´Arts & Metiers (ENSAM), Hassan II Univ. - Mohammedia, Casablanca, Morocco
  • fYear
    2014
  • fDate
    20-22 Oct. 2014
  • Firstpage
    306
  • Lastpage
    310
  • Abstract
    In this paper, we present an exhaustive database for Modern Standard Arabic (MSA) verbs generated from trilateral roots. This database is initially represented as a root-pattern matrix listing rows of all recognized roots and columns of all verb patterns in MSA. The intersection of each row and column contains an index indicating the compatibility of the aforementioned root-pattern pair. This index refers also to a list of morpho-syntactic characteristics of the generated verb. We later converted the database into the more flexible XML format. The aim for our approach is twofold: with the objective of building an exhaustive list, we opted for automatic generation of all possible trilateral roots in the Arabic alphabet and subsequent filtering of roots not recognized in the literature; secondly, converting the database into XML creates a highly versatile resource for easy integration in Arabic NLP applications.
  • Keywords
    XML; database management systems; information filtering; natural language processing; text analysis; Arabic NLP applications; Arabic alphabet; MSA verbs; XML database; exhaustive database; exhaustive list; flexible XML format; modern standard Arabic verbs; morpho-syntactic characteristics; root-pattern matrix; root-pattern pair; roots filtering; triliteral roots; verb patterns; Buildings; Filtering; Indexes; Pragmatics; Standards; XML; Arabic NLP; XML linguistic resources; lexical database; matrix root-pattern; morphosyntax;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Information Science and Technology (CIST), 2014 Third IEEE International Colloquium in
  • Conference_Location
    Tetouan
  • Print_ISBN
    978-1-4799-5978-5
  • Type

    conf

  • DOI
    10.1109/CIST.2014.7016637
  • Filename
    7016637