Frequent Pattern Reorganization Approach to HTTP Botnet Detection

Abstract

Botnet is the most frightful and malicious threats that usually occurs in the current cyber security. It can be
defined as collection of compromised computers that are controlled remotely by hackers or botma ster. Malicious
activities of botnet such as launching DDoS (distributed denial of service)attacks, sending spam, Trojan, Phishing
emails, information harvesting and click fraud. Recently malicious botnets progresses into HTTP botnets out of typical
IRC and P2P botnets. The defining characteristics of botnet are the use of command and control channel through which
they can be updated and directed. Data mining techniques and algorithms allow us to automate detecting characteristics
from large amount of data, which the conventional heuristics and signature based methods could not apply. One major
data mining technique for extracting valuable pattern of the botnet attack is FP-growth algorithm. It aims to design and
implement mechanism to detect bot activity. After perform proposed approach we can discovers regularities and
irregularities in large amount of data set.