Enhancing quality Arabic web content through multilingual information retrieval methods

مقدم أطروحة جامعية

Safa, Abd al-Fattah

مشرف أطروحة جامعية

Yahya, Adnan

الجامعة

جامعة بيرزيت

الكلية

كلية الهندسة و التكنولوجيا

القسم الأكاديمي

دائرة علم الحاسوب

دولة الجامعة

فلسطين (الضفة الغربية)

الدرجة العلمية

ماجستير

تاريخ الدرجة العلمية

2019

الملخص الإنجليزي

Access to quality web content is an integral part of modern life.

While Arabic content is increasing rapidly, it is still far from adequate in terms of quantity, quality and timeliness and is well below what is available in many other languages.

To meet user needs, it helps to give Arabic users access to foreign web content.

The focus of the thesis is to grant Arabic users access to quality web content in other languages to meet their information needs.

This includes access to: textual data, structured databases and annotated multimedia elements.

For that we rely on Cross Lingual Information Retrieval (CLIR) concepts such as cross lingual document similarity, relevance measures, named entity recognition/transliteration, document quality assessment tools and knowledge extraction results [1,2].

Among the approaches we build on are Semantic Association (Explicit and Latent) [1,3,4], Knowledge Extraction [5,6,7] and Machine Translation and Transliteration [8].The methods developed will be utilized to increase the Arabic web content, to suggest material for further processing using crowd sourcing and, as a byproduct, for cross lingual plagiarism detection.

التخصصات الرئيسية

تكنولوجيا المعلومات وعلم الحاسوب

عدد الصفحات

153

قائمة المحتويات

• APA
• MLA
• AMA

لغة النص

الإنجليزية

نوع البيانات

رسائل جامعية

رقم السجل

BIM-1413053