This paper explores the growing threat of phishing, which continues to target individuals and organizations using increasingly sophisticated tactics. The primary focus is on understanding how phishing has evolved over time, the various forms it takes, and how developing machine learning technologies might assist in detecting and preventing these attacks. Understanding the early strategies of phishing, which began in the mid-1990s, helps explain its progression into the diverse approaches found today, from basic false emails to sophisticated spear phishing, smishing, and deepfake-based vishing.
The study provides a historical overview of phishing, categorizes its different types, and explains the motivations behind these attacks, which often exploit human trust and fear. Both technical and psychological factors that lead to phishing 'success' are discussed, emphasizing the importance of combining technological defenses with user education. Detection strategies are explored, such as machine learning-based email filtering and advanced URL analysis, which are essential tools for mitigating phishing risks. Additionally, machine learning models are evaluated, such as supervised and unsupervised algorithms, which are also effective at phishing detection by analyzing email content, sender behavior, and link credibility.
The conclusions stress that while technological solutions are vital, user education is an irreplaceable component of an effective defense strategy. The paper also examines how global events, such as the COVID-19 pandemic, have influenced phishing, with attackers taking advantage of public uncertainty to boost their chances of success. By investigating current trends, such as the use of AI and behavioral analytics in phishing detection, this study underscores the need for continuous adaptation in cybersecurity to counter increasingly sophisticated attacks. The goal is to provide clear knowledge of phishing's development and evolution and urge proactive defenses to minimize its impact, protecting both individuals and organizations from this ongoing threat.
The Evolution of Phishing: Types, Threats, and Machine Learning Mitigation Strategies
Category
Student Abstract Submission