The student Jamal Helal Asi received his PhD degree from Computer Sciences department / University of Technology for his thesis titled (Intrusion Detection Analysis using Machine Learning and Improved Artificial Immune System). In this thesis, the Dataset was reduced from 41 to 10 important properties, the accuracy was high (84.1509%) and the time was (6.11 seconds) with the classification algorithm (J48). The second proposal is the AIRS algorithm, , The proposed system is the VNS classification. The principle of this algorithm is the successive detection of the neighbors to find the best solutions and produce a set of real vectors ready for classification. The results are based on NSL-KDD dataset and showed an increase in performance accuracy compared to the original algorithm (AIRS). The result was (88.136% 9872.15 seconds). In the third proposal, AIRS is an advanced algorithm developed by AIRS. It was suggested to improve the performance of the AIRS input and copy cells, which would give the best cells that can be compared to patterns in the data set.

This was on Sunday, September 30, 2018, and on the discussion room in the annex building of the department. The discussion committee was composed of: Dr. Soukaina Hassan Hashim, Dr. Alaa Kazem Farhan, (From the University of Technology / Computer Science Department and Prof. Dr. Mahmoud Khalil Ibrahim) from the University of Nahrain / Information Engineering and (Dr. Nada Abdel-Zahra Abdullah) from the University of Baghdad / Faculty of Science / Department of Computer Science In the presence of the student supervisor (Prof. Ahmed Tariq Sadiq).

Download Abstract

The student (Hajar Najah Abdel-Saheb) obtained a master's degree from the Department of Computer Science / University of Technology for her thesis (Hybrid Biometric Identification System Based on Principal Component Analysis (PCA) Features Extraction And Neural Network (NN) Classifier) Among several people whose biometrics were collected in a database. The vital features that we rely on in this message are teeth and DNA. Basic concepts are introduced to identify one of many through the use of biometric standards and how these standards have been improved in the last few years. It presents various research and techniques related to this subject and concepts for evaluating these techniques. Presents a new way to identify a person from his own DNA and teeth which is widely used in forensic, border regulation, paternity determination and in investigations. The system contains three main modules: the dental model, the DNA model and the dental model mixture and the DNA (hybrid). The dental model has two main stages: the training phase (sampling of the database) and the testing stage (identification). These stages include the processing stage and the stage of extracting the characteristic features.

This was on Thursday 27/9/2018 at the discussion hall in the annex building of the department. The discussion committee consisted of (Dr. Soukaina Hussein Hashim and Dr. Rahim Abdel-Saheb Ogla) from the University of Technology / Department of Computer Sciences and (Dr. Samer Saeed Ibrahim) from Al Rafidain College / Department of Computer Science in the presence of the student's supervisor (Dr. Shaimaa' Hamid Shaker).

Download Abstract