Семинар за вештачку интелигенцију, 20. децембар 2023.

• 15. Децембар, 2023
• Коментари (0)

Наредни састанак Семинара биће одржан онлајн у среду, 20. децембра 2023. са почетком у 19 часова.
 
Предавач: Ђорђе Јовановић, Математички институт САНУ
 
Наслов предавања: DETECTION OF BOTNET CnC PATTERNS USING TIME SERIES CHARACTERISTICS
 
Апстракт:
Botnets are a class of viruses which use a network of infected computers, connected via internet, to execute a DDoS attack against a server. With the emergence of Mirai botnet, this class of viruses gained worldwide attention from both industry experts and academia. Since the entirety of botnet communication occurs via computer networks, header inspection and deep packet analysis are two most common methods of botnet detection. In this lecture a third method of botnet detection, which focuses on time series characteristics, will be used to extract features of botnet CnC communication, in an attempt to properly detect this type of malware, and, ultimately, mitigate the damage it might make. The dataset for this research used captured network traces of RaspberryPi devices, which have been infected with newest samples of Mirai and Gafgyt viruses. Also, special attention shall be directed towards various particularities of such a problem, such as handling class imbalance, machine learning pipeline, hyperparameter optimization.
 
Напомена: Регистрациона форма за учешће на Семинару је доступна на линку:
https://miteam.mi.sanu.ac.rs/asset/CW5nJWDSEZDj7p32p

Уколико желите само да пратите предавање без могућности активног учешћа, пренос је доступан на линку:
https://miteam.mi.sanu.ac.rs/asset/4LNW8WtML7rLKojoz