Innovations of phishing defense: The mechanism, measurement and defense strategies

Abstract

Now-a-days, social engineering is considered to be one of the most overwhelming threats in the field of cyber security. Social engineers, who deceive people by using their personal appeal through cunning communication, do not rely on finding the vulnerabilities to break into the cyberspace as traditional hackers. Instead, they make shifty communication with the victims that often enable them to gain confidential information like their credentials to compromise cyber security. Phishing attack has become one of the most commonly used social engineering methods in daily life. Since the attacker does not rely on technical vulnerabilities, social engineering, especially phishing attacks cannot be tackled using cyber security tools like firewalls, IDSs (Intrusion Detection Systems), etc. What is more, the increased popularity of the social media has further complicated the problem by availing abundance of information that can be used against the victims. The objective of this paper is to propose a new framework that characterizes the behavior of the phishing attack, and a comprehensive model for describing awareness, measurement and defense of phishing based attacks. To be specific, we propose a hybrid multi-layer model using Natural Language Processing (NLP) techniques for defending against phishing attacks. The model enables a new prospect in detection of a potential attacker trying to manipulate the victim for revealing confidential information.