Research Projects

a.    "Context-based morphological analyzer" Project, which will start very soon. Morphological analysis techniques     analyze natural words by considering their internal morphological structures. Available morphological analyzers, such as  AraMorph for example, return all possible solutions of the analyzed word without considering its context if it is provided in a sentence. Our objective in this project is to build a statistical model based on the Arabic sentence structure for selecting a best solution from those returned by a morphological analyzer using the context. It is funded by the Dean of Scientific Research at the Imam University.

b.    "Virtual Translator from Arabic Text to Sign Language" Project. Two main problems are being treated in this project, automatic translation from Arabic to sign language (machine translation problem) and visualization of signs (computer graphics problem). A statistical approach based on the sentence structure of both source and target languages will be considered for the first problem. For the second one, an avatar approach (3D rendering) is under consideration by developing signing system with two components: sign-builder and sign-player. At the end of the project, an educational tool based on these approaches will be developed as a web-application for teaching Arabic sign language. This project is funded by the Science & Technology Unit at the Imam University, which is created in the framework of the national plan for Science, Technology and Innovation (STI), supervised by King Abdulaziz City for Science and Technology (KACST). It is for two years starting from January 2010.

c.    A book project entitled "Human and Computer Interaction: concepts and techniques (in Arabic)", funded by the Deanship of Scientific Research at Al-Imam University since December 2008 (for two years).

d.    "Computerized Teaching of the Holly Quran" Project, which aims to develop an appropriate environment for self learning of the Holy Quran and its sciences. Different tracks have been designed to carry out this project. In the first track, the focus was on gathering and improving the most important aspects related to the recitation and memorization of the Holy Quran that are available in the existing Quranic programs, and then putting them in a comprehensive and consistent environment. In the second track, speech recognition techniques have been used to assist reading the noble Quran. In the third track, techniques for determining the similarity between verses (ayah آية) of the noble Quran were investigated. Computer-tools have been developed for analyzing the text of the noble Quran based on complete words and their stems in order to link similar verses. In the fourth track, a sub-system for teaching tajweed rules was developed. A mathematical formulation of tajweed rules was proposed and then programmed in an engine that can be used to detect places of tajweed rules in the Quranic verses and also to assist learning them. This project was sponsored by KACST for four years, starting from March 2006 (grant number AT25-113). It is almost finished and has led to very important outcomes: corpora for the holly Quran (sound corpus and textual one), HMM-based recognizer for the Quranic sounds, engine for Tajweed rules, another engine for similarity between Quranic terms, etc.

e.    "Part-Of-Speech tagging (POS) For Arabic Language" Project, which aims to use an appropriate stochastic approach with a rule-based one to build a model able to identify the main constituents of a given Arabic sentence. For this purpose, we have used a hybrid method that combines morphological analysis with Hidden Markov Models (HMM) using the Arabic sentence structure. On the one hand, the morphological analysis is used to reduce the size of the tags lexicon by segmenting Arabic words in their prefixes, stems, and suffixes due to the fact that Arabic is a derivational language. On the other hand, HMM is used to represent the Arabic sentence structure in order to take into account the logical linguistic sequencing. An appropriate tagging system has been proposed to represent the main Arabic part of speech in a hierarchical manner allowing an easy expansion whenever it is needed. Each tag in this system is used to represent a possible state of the HMM and the transitions between tags (states) are governed by the syntax of the sentence. This project is also sponsored by KACST as a part of a big project relating to the development of Arabic tools processing. It has been successfully finished and is currently used in our research to tag the holly Quran and some classical Arabic texts.

Additional information