Abstract

We live in a time when data security has become a significant concern. Cyber services are the most enjoyable and time-saving aspects of one's life. On the other hand, people save their data in the cloud, handled by the cyber. In this case, cyber-security is quite vital. This is an open security challenge because many intruders can attack the data and hack the user's details through the server. If we look around, we will see a lot of cases involving cyber-crime. The security of our cloud-based datasets has become a serious concern. Our research will include data security, including intruder detection that can happen anywhere on the planet. Protecting data from intruders has become critical, and intruder detection should be the essential key to identifying. How will we know who is stealing the data that has been secured using biometric security, fingerprints, passwords, OTPs, and other methods if we don't know who the intruder is? Intruder detection has become increasingly important, particularly on mobile objects such as aeroplanes and ships. We can only find a solution if we understand the problem. We employ machine learning, biometric recognition, data learning, and hybrid approaches to avoid this. These will be the system's handles, and they will help secure data from intruders by utilizing the best optimization techniques to obtain precise data. We proposed a banking system model in which biometric impressions and digital signatures are used to enable every transaction by a bank's customer. This proposal recommends that the Smart Online Banking System (SOBS) be made more secure by employing biometric prints, which decreases the number of threats that an invader may pose.