Implementation of a Scalable and Secure Cloud Framework for Heart Disease Classification Using a Hybrid SVM-Neural Network Model with Congestion Control

Abstract

To adequately protect our data on the cloud, we rely on strong security measures designed to discourage intruders. One key tactic we use is classification, a supervised learning method in cloud security that helps determine the group affiliations of data instances. Additionally, we have implemented a congestion management system that limits access rights and enhances user identity protection using artificial intelligence. Our technology proactively detects any fraudulent activity and alerts the user, eliminating the need for standard verification methods like passwords or OTPs. Instead, it evaluates risk by taking into account device location, network conditions, and user behavior. Once hackers have login credentials, sophisticated detection techniques are employed to identify unusual activities. To guarantee that users can function dependably and securely, some systems may restrict access or require further verification. We could need extra security measures, such as biometric verification or special passwords, in high-risk situations. We'll also look at a hybrid classification approach that combines the advantages of several different categorization methods. This research will look at a number of widely used methods for categorizing medical diagnosis data, with an emphasis on a hybrid model that blends neural networks with support vector machines. Additionally, we will compare the accuracy rates of these algorithms across many categories and assess their effectiveness using a range of metrics.