Arabic document retrieval based on word rooting

مقدم أطروحة جامعية

al-Sarhan, Hind Hammud

مشرف أطروحة جامعية

al-Rababiah, Mamun

الجامعة

جامعة آل البيت

الكلية

كلية الأمير الحسين بن عبد الله لتكنولوجيا المعلومات

القسم الأكاديمي

قسم علوم الحاسوب

دولة الجامعة

الأردن

الدرجة العلمية

ماجستير

تاريخ الدرجة العلمية

2013

الملخص الإنجليزي

In this research, an Arabic morphological system designed and implemented to extract Arabic roots.

Patterns play a big role in a process of extracting the root of the word, each word has its own pattern as in "يتفكرون ،يتفعلون " each letter in the word that replaces the letters ف،ع،ل in the pattern is considered as original letter and added to the root.

Efficient results are gotten that provide a good step in the process of minimizing the size of index structure.

The system was designed to accept input from a file or it can analyze one word at a time.

Words that the algorithm fails to analyze are normally foreign words, irregular words or words that do not have trilateral roots.

The new algorithms used in the proposed morphological analyzer are minimal in searching time.

The proposed system is directed to the modern Arabic.

It can resolve the inflected Arabic word, identify its root and finds its pattern.

Small components can be added to this system so that it will be capable of finding the verb tense and many other word attributes. This can be done using the information deduced by the proposed system and some extra linguistic rules.

التخصصات الرئيسية

تكنولوجيا المعلومات وعلم الحاسوب

الموضوعات

• معالجة البيانات
• اللسانيات الحاسوبية
• اللغة العربية

عدد الصفحات

66

قائمة المحتويات

• APA
• MLA
• AMA

لغة النص

الإنجليزية

نوع البيانات

رسائل جامعية

رقم السجل

BIM-414592