Cybersecurity Framework Development for BoTNet Attack Detection using ISSOA based Attention Recurrent Autoencoder

These days, most people's time is devoted to various online communities that provide a wide variety of personal and professional services as well as social media platforms. Hence, it is crucial to safeguard these environments against cybercriminals who may cause data theft or system disruptions. To safeguard consumers from intrusions and threats, keep personal data private, and keep information systems running smoothly, cybersecurity measures include technical, organizational, and executive measures to prevent unauthorized access to, or misuse of, information and communication systems. In light of this, the essay delves into the importance of AI in cybersecurity as well as the procedures that safeguard computer systems from assaults, hacking, and data theft. The article suggests an enhanced sparrow search optimization algorithm (ISSOA) for feature selection. In the end, the Adamax optimizer with attention recurrent autoencoder (ARAE) perfect is used to categorize the attacks. The realistic-network-traffic BoT-IoT dataset, which is open to the public, is used for the tests. By predicting and comprehending the behavior besides traffic of malicious software, machine learning besides deep learning algorithms significantly contribute to the management of system infiltration and the protection of computer systems from unauthorized entry. Particularly impressive are the suggested classifier's 97% and 98% recall accuracy rates.