ارسل ملاحظاتك

ارسل ملاحظاتك لنا







الرئيسية > بحث>Translating Natural Language C...>الوصف

Translating Natural Language Constraints To OCL

المصدر: مجلة جامعة الملك سعود - علوم الحاسب والمعلومات
الناشر: جامعة الملك سعود
المؤلف الرئيسي: Bajwa, Imran Sarwar (Author)
مؤلفين آخرين: Lee, Mark (Co-Author), Bordbar, Behzad (Co-Author)
المجلد/العدد: مج24, ع2
محكمة: نعم
الدولة: السعودية
التاريخ الميلادي: 2012
الصفحات: 117 - 128
DOI: 10.33948/0584-024-002-004
ISSN: 1319-1578
رقم MD: 972799
نوع المحتوى: بحوث ومقالات
اللغة: الإنجليزية
قواعد المعلومات: science
مواضيع:
اللسانيات الحاسوبية | اللغة الإنجليزية | برامج الحاسوب
كلمات المؤلف المفتاحية:
Natural Language Processing | English Constraints | Formal Constraints
رابط المحتوى:
PDF (صورة)     
صورة الغلاف QR قانون
حفظ في:
المستخلص: Object Constraint Language (OCL) is the only available language to annotate the Unified Modeling Language (UML) based conceptual schema (CS) of a software application. In practice, the constraints are captured in a natural language (NL) such as English and then an OCL expert manually transforms the NL expressions to OCL syntax. However, it is a common knowledge that OCL is difficult to write specifically for the novel users with little or no prior knowledge of OCL. In recent times, model transformation technology has made transformation of one language to another simpler and easier. In this paper we present a novel approach to automatically transform NL specification of software constraints to OCL constraints. In NL to OCL transformation, Semantics of Business Vocabulary and Rules (SBVR) standard is used as an intermediate representation due to a couple of reasons: first of all, SBVR is based on higher order logic that simplifies the transformation of SBVR to other formal languages such as OCL. Moreover, SBVR used syntax of natural language and thus is close to human beings. The presented NL to OCL transformation via SBVR will not only simplify the process of generating OCL constraints but also generate accurate models in less time.
ISSN: 1319-1578

عناصر مشابهة


© 2024 المنظومة. جميع الحقوق محفوظة.