A Cloud-Assisted Framework Utilizing Blockchain, Machine Learning, and Artificial Intelligence to Countermeasure Phishing Attacks in Smart Cities

Phishing attacks are a major Cybersecurity threat, specially in smart cities. In recent years, there has been a growing trend of phishing attacks targeting smart city infrastructure. These attacks can have a significant impact on the safety and security of smart cities. This paper presents a cloud-assisted framework for countering phishing attacks in smart cities. The framework uses a combination of machine learning and blockchain technologies to detect and prevent phishing attacks. The framework was evaluated using a dataset of phishing emails and was shown to be effective in detecting phishing attacks with high accuracy. Experimental results demonstrate the framework's effectiveness in detecting and blocking phishing attacks, providing accurate and timely responses. Moreover, the framework offers cost-efficiency in terms of implementation and maintenance. Evaluation metrics encompass the number of successfully detected and blocked attacks, the efficiency of the detection and prevention process, the accuracy of the machine learning and artificial intelligence models, and cost considerations. The quantitative results of the evaluation showed that the framework performed well in countering phishing attacks in smart cities. The accuracy ranged from 0.92 to 0.95, the precision scores ranged from 0.91 to 0.94, the recall rates ranged from 0.93 to 0.96, and the F1 score ranged from 0.92 to 0.95. The false positive rates ranged from 0.09 to 0.05, and the false negative rates ranged from 0.07 to 0.04. The true positive rates ranged from 0.93 to 0.96, and the true negative rates ranged from 0.91 to 0.94. The area under the ROC curve (AUC) ranged from 0.95 to 0.97. The framework demonstrated low training times of 30 to 60 seconds and fast inference times of 5 to 10 milliseconds. Resource utilization ranged from 80% to 75%. The framework exhibits high scalability and robustness. Making it suitable for deployment in real-world environments.