The student Sara Mohammed Sherif received her master's degree from the Department of Computer Sciences / University of Technology for her thesis (A Proposed Algorithm to Decrease False Alarm in Intrusion Detection System) the thesis presented a proposed algorithm for deep analysis of the network intrusion detection system Which is a combination of various data extraction techniques for accurate diagnosis of attacks and with the least false alarm rate. The proposed system involves the use of three mining techniques on three consecutive levels: at the first level, the simple hypothetical NB algorithm is used to detect abnormal activity from normal behavior. In the second level, the logistic regression algorithm is used to classify abnormal activity into the four main attack types To the natural activity category. Finally, at the third level, ID3 resolution tree uses an algorithm to classify four major attack types into (22) intrusion attacks in addition to normal activity. To evaluate the motion system KDDCUP99 uses the standard data set for intrusion detection.

The proposed system is based on the application of the processing phase, which includes data conversion and the technique of selecting the most important characteristics, in order to increase the accuracy of the system and reduce the time. This was on Thursday 19/4/2018 and on the discussion room in the department. The discussion committee consisted of Prof. Dr. Abdel Moneim Saleh Rahma and Dr. Rahab Falih Hassan of the University of Technology / Computer Sciences Department and Dr. Haidar Kazem Hamoud of Al Mustansiriya University / Faculty of Education. / Department of Computer Science in the presence of the supervisor of the student (Dr Soukaina Hassan Hashim).

Download Abstract