Research

Interests

  • Data Mining
  • Information Retrieval
  • Machine Learning
  • Artificial Intelligence
  • Expert Systems
  • Network and Data Security
  • Decision Support Systems

Research Projects

  • Network Intrusion Detection System Based on Normal behaviour

    Network Intrusion Detection System Based on Normal behaviour

    we proposed OCC-NIDS model based on the standard deviation of service’s normal behaviour. Through this model we dealt with each network service as single class instead of dealing with all network services as a single class.

    A lot of efforts have been given toward designing a perfect NIDS that has a high detection rate and low false alarm rate. Some have used misuse detection technique which fails to detect zero-day attacks, while the problem of using supervised learning is the cost of producing labeled dataset which is essential for training the model and also the model is trained on known attacks which may fail to detect new variant attacks. On the other hand, unsupervised learning has the problem of labeling the generated clusters. Once-Class Classification learning technique (OCC) suffers from the high dimensional network feature spaces, Also, problems may arise when large differences in density exist. To overcome these problems, we proposed OCC-NIDS model based on the standard deviation of service’s normal behaviour. Through this model we dealt with each network service as single class instead of dealing with all network services as a single class. By this way we use just the relevant features of each service, hence reducing the high dimensional network feature spaces and also ensure that each class has – a proximately – uniform distribution. The proposed model proved that it is able to detect abnormal network traffic with high detection rate and low false positive rate. It achieved 99.72% detection rate and 99.65% accuracy rate with a false alarm rate reached 0.7% and false positive rate 0.005% on KDD Cup’99 dataset.

  • Implementing a Multi-threaded FP-Growth algorithm

    Implementing a Multi-threaded FP-Growth algorithm

    This research examine the performance of FP-Growth and Multi-threaded FP-Growth algorithms by comparing their time complexity

    Testing the FP-growth algorithm using the multi threading to measure the time it takes and comparing it with the FP-growth without multithreading, figure-5 shows this using connect data set, we see that the FP-growth with multithreading take minimum time in generating the frequent item sets than the FP-growth without multithreading takes with the same support.

  • Mining Educational Data to Improve Students’ Performance In General Physics1, A Case Study

    Mining Educational Data to Improve Students’ Performance In General Physics1, A Case Study

    In this paper we used educational data mining to improve students’ performance in Physics course, to overcome the problem of low grades, and dropouts in this course. In our case study we try to extract useful knowledge from students historical data collected from the Al-Aqsa University – Gaza

    We observed, based on classification and clustering, that the course General Physics1 needs a hard work, this is why the students with high GPA gets high grades in this course. We observed also that most of the excellent and very good grades in General Physics 1 are belongs to students who attended at Mathematics Department and because General Physics1 needs a good background in Principles of Mathematics, any student who studied the course Principles of Mathematics course have a high chance to  success course General Physics1.

  • Continuous Evaluation and Monitoring of Web Service’s QoS Dynamic Attributes for each of its binding protocols

    Continuous Evaluation and Monitoring of Web Service’s QoS Dynamic Attributes for each of its binding protocols

    In this paper we propose a new mechanism for monitoring and evaluation web service’s QoS dynamic attributes to rank each of its binding protocol for the published web service

    Automated Web service discovery based on Quality of Service (QoS) parameters is a key challenge in Service Oriented Architecture (SOA) research. There exist many web services which exhibit similar functional characteristics. It is imperative to provide service consumers with facilities for discovering required web services according to their nonfunctional aspects like QoS. The trustworthiness of published QoS information and reputation for Web service discovery has been a challenging issue for the operation of selecting the suitable web service according to the user’s needs.
    In addition, in an untrustworthy real world environment, the QoS-based service discovery approach cannot verify the correctness of web services’ QoS, since such values guaranteed by a service provider are different from the real ones. In fact invoking a low quality web services can affect the overall performance of the composite web services, besides it’s a challenging issue in the area of dynamic Web service composition