Malicious website identification using design attribute learning

Malicious websites pose a challenging cybersecurity threat. Traditional tools for detecting malicious rely heavily on industry-specific domain knowledge, are maintained by large-scale research operations, and result in never-ending attacker–defender dynamic. need to balance two opposing requirements successfully function: escaping malware detection while attracting visitors. This fundamental conflict can be leveraged create robust sustainable approach based the extraction, analysis, learning of design attributes website identification. In this paper, we propose next-generation algorithm extended attribute that learns analyzes web page structures, content, appearances, reputation detect websites. Results from experiment was conducted more than 35,000 suggest proposed effectively detects 83% all maintaining low false-positive rate 2%. addition, method incorporate user feedback flag new suspicious thus effective against zero-day attacks.