Researches on detection of Anomalies in network traffic conducted

The next scientific seminar of Department № 1 was held at the Institute of Information Technology of ANAS.  The chief engineer of the Institute, PhD in technical sciences, associate professor  Makrufa Hajirahimova presented the report on " detection of anomaly in network traffic using the classification ensemble".  The reporter noted that recently the number of cyber-attacks has increased rapidly and network services and web add-ons have made it necessary to implement security and security measures against network and computer threats in enterprises and organizations.

According to her, Collection of large volumes of data from network, host, security devices, etc. sources to detect attacks that threaten the security of corporate computer networks updated the issue of detecting anomalies in the data and there is a need for more effective analysis methods and algorithms for the solution of the problem.

The reporter informed about the most commonly used threat detection systems (Intrusion Detection System - IDS) in the detection of information threats, Security information and event management (SIEM), as well as, II generation SIEM systems (Predictive Big Data Security.

She also informed about stochastic and determinative methods of detection of network anomalies and existing approaches. She noted that in many studies, hybrid or multilevel classification models are proposed to increase the accuracy of classification in the detection of anomaly.

Provided an overview of the multi-classifier model (J48, LogitBoost, IBk, AdaBoost, RandomTree), that provides a high-precision detection of malicious traffic in the network to increase the efficiency of IDS's work, the stages of the model development, the selection of training and test sets (NSL-KDD), the classification efficiency metrics and the classification accuracy of the classifiers.

Then, the employee of the department, Gulendam Nuriyeva, spoke on "MATLAB application software".  The speaker first spoke about the history of MATLAB software and stated that MATLAB, a shortening of "Matrix Laboratory", was written in Fortran in 1970 by Cleve Moler.  Later, Moler re-worked with Jack Little and Steve Bangert in MATLAB C and set up MathWorks Inc. in 1984. 

The reporter noted that MATLAB allows to work with  matrix manipulations, layout of function and data graphs, algorithms implementation , user interface development, Fortran, C, C ++, Java and other   programs written in programming languages.

"For engineers and researchers, MATLAB is the easiest and most productive environment," said the reporter and emphasized that this program was specifically designed for engineers and researchers. Additionally, MATLAB, which is rich in functionality and libraries, allows to use applications without any code.

Finally, discussions were held around the reports, and questions were answered.

© All rights reserved. Citing to www.ict.az is necessary upon using news