أبحاث أعضاء الهيئة الأكاديمية 2012

  • # -----> #

1. Tawfiq Barhoom and Hanaa Qeshta
"Worm Detection by Combination of Classification With Neural Networks "  
Abstract Security has become ubiquitous in every domain today as newly emerging malware pose an ever increasing perilous threat to systems. Worms are on the top of malware threats attacking computer system although of the evolution of worm's detection techniques. Early detection of unknown worms is still a problem. In this paper, we proposed and implemented a new approach for worm's detection. The proposed model uses combination of Decision Tree and Neural Network (NN) as classifying worm/ non worm traffic network. Our results showed  that the detection rates of classification and detection known worms are at least 93.51% with NN, and 92.87% with Decision Tree, while the unknown worm detection rates was about 97.27%  with NN , and  93.2%  with Decision Tree. The detection rate of our proposed model in known worm was 95.59% while the unknown worms detection rate was 97.74%.

2. Abdelkareem Alashqar
"A comparative study on Arabic POS tagging using Quran corpus ", The 8th International Conference on Informatics and Systems (INFOS 2012), Cairo University, Cairo, Egypt.
Abstract POS tagging is the process of computationally assigning correct part of speech to each word of a given input text depending on the context. Different POS tagging techniques in the literature have been developed and experimented mostly for English language. Some of the same work has been done for Arabic language. Comparative studies on POS tagging for Arabic language are relatively unexplored.  In  this  paper  we  compare  the  performance  of  some POS  tagging  techniques for Arabic  using  Quran corpus. These techniques include N-Gram, Brill, HMM, and TnT taggers. The comparison experiments have been done on diacritized and undiacritized classical Arabic.  We tried to see which technique maximizes the performance with our case. 

1. Ayad, F. and Alashqar, A.
"Effectiveness of Using Web 2.0 Tools in Learning Management System (Moodle) for Achieving Collaborative Learning among Information Technology Students at the Islamic University", Information Studies Journal, Vol. 10.  
Abstract The research aimed at identifying the characteristics that must be available for the wiki tool in Moodle, and investigating the importance degree of these characteristics, and the benefit degree from these characteristics among information technology students at the Islamic University. Descriptive approach was used, and the research tool consisted of a questionnaire for evaluating the importance degree of wiki characteristics, and the benefit degree from these characteristics. The questionnaire consisted of four main categories: wiki and interaction among students, wiki and interaction of students with teacher, wiki and interaction of students with resources, and wiki and interaction with wiki program services. The research sample consisted of all students who registered for information systems course at the Islamic University of Gaza in the second semester 2009-2010, where the number of male students was (23), and the number of female students was (21). The results showed a high degree of importance of the characteristics of  the wiki tool among the students of the research sample, and collaborative  learning among these students have been well achieved through using the  wiki tool. The results also showed that there were no statistically significant differences between male and female students on the importance and benefit degrees. The researchers recommended encouraging teachers to employ interaction and collaborative learning tools especially the wiki.

1. Tawfiq Barhoom and Hanaa Qeshta
"Worm Detection by Combination of Classification With Neural Networks "  
Abstract Security has become ubiquitous in every domain today as newly emerging malware pose an ever increasing perilous threat to systems. Worms are on the top of malware threats attacking computer system although of the evolution of worm's detection techniques. Early detection of unknown worms is still a problem. In this paper, we proposed and implemented a new approach for worm's detection. The proposed model uses combination of Decision Tree and Neural Network (NN) as classifying worm/ non worm traffic network. Our results showed  that the detection rates of classification and detection known worms are at least 93.51% with NN, and 92.87% with Decision Tree, while the unknown worm detection rates was about 97.27%  with NN , and  93.2%  with Decision Tree. The detection rate of our proposed model in known worm was 95.59% while the unknown worms detection rate was 97.74%.