The Whalebone CTO Robert Sefr has been selected as one of the speakers for this year's QuBit IT security conference in Belgrade. Using Neural Networks to detect malicious domains is a result of long term joint research with Prague Technical university and Whalebone is very excited about the results of the works!Pattern recognition is one of the common use cases of Neural Networks. Deep learning brings lot of new possibilities in detection previously unknown attacks, however could be also tricky because of the false positives. Speaker will present fresh results of almost year lasting research in the area of detecting threats in DNS traffic. The research team was put together from specialists from Whalebone and researchers from Czech University of Technology with extensive experience in deep learning and botnet detection. During the presentation two approaches will be described. One is a Neural network detecting randomly generated domains (aiming at Domain Generation Algorithm based botnets), the second one aims at behavior of clients over time (aiming more at the detection of infected machines externalities). Speech will go into detail about results from the real world DNS traffic originating from regular households and enterprises.See you in Belgrade at QuBit 2017!
