A Network For Collaborative Detection Of Intrusions In Smart Cities Using Blockchain Technology

Abstract

The field of cybersecurity has undergone significant transformation with the integration of machine learning (ML) and artificial intelligence (AI) techniques into intrusion detection systems (IDS). This research article presents a comprehensive survey spanning the past five years, exploring the symbiotic relationship between ML, AI, and intrusion detection. The survey traverses seminal studies, methodologies, and results, shedding light on an evolving landscape characterized by innovation and advancement. The classification report’s key metrics—precision, recall, F1-score, and support. High precision values point to accurate positive predictions, while recall values showcase the model’s ability to capture true instances. The F1-score signifies the equilibrium between precision and recall. These metrics collectively underscore the model’s proficiency in identifying and differentiating intrusion classes, reinforcing its real-world applicability. In conclusion, this research article presents a holistic view of ML and AI integration with intrusion detection, offering insights into innovative contributions and their implications for cybersecurity. While highlighting existing research gaps, the article underscores the potential of AI-driven intrusion detection systems and advocates for ongoing advancements to fortify digital security against emerging threats.