Social Engineering Threat: Phishing Detection using Machine Learning Approach

In cyberspace, social engineering attacks are common, well-known, and simple to use. The overwhelming majority of hostile attempts against both physical and virtual IT systems were founded on or started utilizing social engineering techniques, according to historical evidence of such attacks. It is conceivable that the lockdown had an impact on crime rates because COVID-19 altered the opportunity for crime. This implies that trends in crime-related activities would likewise be impacted by the societal changes brought on by COVID-19. Social engineering attacks come in a variety of forms, such as phishing, pretexting, baiting, and tailgating. In a phishing attack, the victim is asked to enter important information or click on a harmful link in an email or message that appears to be from a reputable source, such a bank or social media site. This paper illustrates the machine learning (ML) approaches for social engineering (SE) assault threat identification. Cybercrime of the phishing variety has increased in recent years. It is a type of social engineering assault where an attacker fools people by sending emails or messages through social media platforms. Phishing attempts either enable the installation of malicious software or steal users' private data. Because attackers can create phishing messages that appear legitimate to a user, they are difficult to spot, and it is a security risk that has negative impacts on both the persons and the brands it targets. The literature provides a wide range of strategies to address this problem, particularly the identification of phishing websites. This paper presents results obtained in each model's data pre-processing, extraction of features, model design and the results are used to evaluate cutting-edge machine learning techniques for identifying URL-based and phishing assaults in depth.