One of the major issues facing the globe now is cyberattacks. Every day, they lead individuals and nations to suffer severe financial losses. Cybercrime is also on the rise along with cyberattacks. The ability to recognize cybercriminals and comprehend their methods of attack is essential in the fight against crime & criminals. Cyber-attacks may be difficult to recognize and prevent. However, scholars have recently developed security systems and made predictions using AI (Artificial Intelligence) techniques to resolve the issues. There are several crime prediction techniques listed in the literature. However, they struggle to forecast the strategies used in cybercrime and cyberattacks. The solution to this issue is to use actual data to identify an assault and its perpetrator. The information includes the sort of crime, the perpetrator's gender, the damage, and the assault techniques. Applications made by people who were the target of cyberattacks may provide the forensic units with the necessary data. In this study, we use Machine Learning (ML) to assess two alternative cybercrime models and project the impact of the stated variables on the identification of the cyberattack technique and the perpetrator. In our methodology, we used eight ML techniques and found that their accuracy rates were comparable. With an accuracy rate of 95.02 percent, the Support Vector Machine (SVM) was determined to be the most effective cyberattack technique. With a high degree of accuracy, the initial model allowed us to forecast the sorts of assaults that the victims were most likely to experience. The most accurate approach for finding attackers was the Logistic Regression, which had a 65.42 percent accuracy rate. In the 2nd model, we forecasted whether a comparison of the perpetrators' traits would allow for their identification. Our findings indicate that the likelihood of a cyberattack diminishes with the victim's level of education and income. We anticipate cybercrime units will implement the recommended approach. It would also ease the identification of cyberattacks and make combating them simpler and more efficient.

Cyber-attack-crimes, Machine learning, Artificial intelligence, Data analysis, Crime prediction, Security and privacy