ارسل ملاحظاتك

ارسل ملاحظاتك لنا







Modeling of an Arabic Electronic Dictionary in Generic Silhouettes for the Embedded Systems

المصدر: وقائع الندوة الثالثة الدولية حول المعالجة الآلية للغة العربية CITALA'09
الناشر: معهد الدراسات والأبحاث للتعريب
المؤلف الرئيسي: Haddad, Ahmed (Author)
مؤلفين آخرين: Ben Ghezala, Hend (Co-Author)
محكمة: نعم
الدولة: المغرب
التاريخ الميلادي: 2009
مكان انعقاد المؤتمر: الرباط
الهيئة المسؤولة: معهد الدراسات والأبحاث للتعريب
الشهر: مايو
الصفحات: 55 - 61
رقم MD: 589498
نوع المحتوى: بحوث المؤتمرات
قواعد المعلومات: AraBase
مواضيع:
رابط المحتوى:
صورة الغلاف QR قانون

عدد مرات التحميل

25

حفظ في:
المستخلص: The objective of this article is to present a system of automatic generation of electronic dictionaries of the classic Arabic language, developed within RIADI labora¬tory (unit in Monastir). This system is part of project "oreillodule": an embedded system of synthesis, translation and recognition of the Arabic word. In this article, we will present the different stages of realization, and notably the automatic generation of these dictionaries based on an original theory: Conditions of Morphemic Structure (CMS), and lexical matrixes. In this case, objectives are bound to calculation time constraints and clutter memory that must be reduced to install the system on machines of reduced capacities (embedded System). That’s why; we designed a dictionary models allowing a strong degree of reduction and precision but little redundancy. At the mod¬eling stage of lexicon, words are grouped according to their describers, which are extracted from their generic silhou¬ettes, forming thus under-lexicons indexed by a describer. To test our approach, we applied it on a system of recogni¬tion of Arabic writing, results are exposed at the end of this article.