Analyzing Spatial Structure of IP Addresses for Detecting Malicious Websites

摘要: Web-based malware attacks have become one of the most serious threats that need to be addressed urgently. Several approaches attracted attention as promising ways detecting such include employing several blacklists. However, these conventional often fail detect new owing versatility malicious websites. Thus, it is difficult maintain up-to-date blacklists with information for To tackle this problem, paper proposes a scheme websites using characteristics IP addresses. Our approach leverages empirical observation addresses are more stable than other metrics URLs and DNS records. While strings form or records highly variable, less i.e., IPv4 address space mapped onto 4-byte strings. In paper, lightweight scalable detection based on machine learning techniques developed evaluated. The aim study not provide single solution effectively detects web-based but develop technique compen- sates drawbacks existing approaches. effectiveness our validated by real data from traffic campus network. results demonstrate can expand coverage/accuracy also unknown covered