Learning models to detect personality traits of cyber attackers: a combined approach using honeypot and surveys.

Abstract

In the current digital age, cyberattacks pose significant threats to global security and privacy, posing complex challenges that society must address. Cybercriminals use increasingly sophisticated techniques to infiltrate systems and extract sensitive information. Understanding and anticipating the behavior of these attackers is essential to develop more agile and efficient defense strategies. This research is methodologically structured in four fundamental phases. Phase 1 consists of information gathering, in which honeypots are deployed to lure attackers and record their activities, providing valuable data on their tactics and behaviors. Phase 2 focuses on experimentation, where this collected data is used to identify initial patterns. Phase 3 employs a descriptive design, integrating survey responses from individuals involved in cyberattack activities to gather direct information about their personalities. Phase 4 involves analyzing the results, applying machine learning models to identify correlations between specific personality traits and cyberattack behaviors. The results of this research contribute significantly to the field of cybersecurity. The developed models demonstrate the ability to predict certain personality traits based on behaviors observed in honeypots and data collected through surveys. These findings expand our understanding of cyber attackers and provide a solid foundation for developing new cybersecurity defense strategies. In conclusion, this work opens new opportunities for interdisciplinary research that integrates psychology and technology in the fight against cyberattacks.