WITHDRAWN: A predictive model of machine learning against phishing attacks and effective defense mechanisms

Abstract

Abstract In recent years, online discontent has become a major threat to the security of information. A phishing email begins the common form of on-line deception, then sends targets to a phishing website where the attacker tries goals to share their credentials. General emails from the phishing targets and the defender are relatively easy to understand. Usually they come from odd addresses, the content is very general and they occur in big quantities. Online deception, however, usually leads to a spear phishing which has significant consequences. Spear-phishing e-mails are sent in a very small volume, target small audiences, sometimes impersonate a trustworthy entity, and redirect targets to a phishing website where an attacker tricks targets to share their identity. Attackers mostly use domain spoof techniques to convince the phishing site even more. The whole process we evaluate. Beginning with phishing mail, we discuss anti-spleen protocols, evaluate the policy and alert of email services and assess the environment for email monitoring. For phishing websites, we use dynamic and static analyses to detect the impersonation of domain names and identify phishing pages. On phishing websites we researched the exchange of credentials. We calculate what happens following the goals. Finally, on modern sites such as Alexa and Google Assistant, we discuss possible questions about phishing and privacy.