الفهرس | Only 14 pages are availabe for public view |
Abstract Intrusion Detection Systems (IDSs), are the most appropriate methods to prevent and detect the attacks of networks and computer systems. The security system development, in the computing world, still requires accurate work. Artificial intelligence technique can make IDSs easier than before. As always, the most important thing is to know more about smart systems through training to acquire the truth things. This thesis focuses on creating an environment for IDSs to teach them to practice the work such as a security officer. The study presents several ways to discover network anomalies using data mining tasks, deep learning technology. In this thesis, two smart hybrid systems were developed to explore any penetrations inside the network. The first model divides into two basic stages. The first stage includes the Genetic Algorithm (GA) in selecting the characteristics with depends on a process of extracting, Discretize And dimensionality reduction through Proportional k-Interval Discretization (PKID) and Fisher Linear Discriminant Analysis (FLDA) respectively. At the end of the first stage combining classifier Naïve Bayes and Decision Table classifier using NSL-KDD data divided into two separate groups for training and testing. The second stage completely depends on the first stage outputs in order to improve the performance in terms of the maximum accuracy in classification of penetrations, raising the average of discovering and reducing of the average of false alarms through participation with the Deep Learning (DL) technology and collaboration with an algorithm (SGD). The second hybrid model relies upon Particle |