Enhanced network intrusion detection system in cyber security for big data applications using machine learning and deep learning algorithms

Abstract

Cyber security has emerged as a top worry for companies as newlinewell as people in modern interconnected society. The environment for danger newlinehas changed significantly as a result of the growing dependence on electronic newlinedevices and the enormous amounts of information being produced. Internet of newlineThings (IoT) adoption and the growth of computer networks bring cyber newlinesecurity issues to light, making it necessary to apply big data analytics and newlinecutting-edge machine learning to anticipate new threats. Modern cyber newlinesecurity concerns are insufficiently addressed by traditional ML techniques. newlineThe IoTis widely used, computer networks are expanding quickly, and there newlineare a ton of relevant applications, so cyber security has recently received a lot newlineof attention in terms of current security concerns. The cyber-universe is newlineexpanding quickly and steadily, which has led to an increase in software newlinedevelopment, data processing, cyber security breaches and the complexity of newlinedefensive tactics. Big data mining methods and cutting-edge machine learning newlinetechniques will be the most effective for use in this challenge, taking into newlineaccount the scale and complexity of the cyber-universe, to forecast brand-new newlineattacks. This is because conventional newlinemachine learning newline ineffective against the cyber security problems of to newline (ML) techniques are newlinereported attacks against IoT systems. The first Stage examines a machine newlinelearning-based IoT-based DoS attack detection. Hence, in this paper, we newlinepropose a --ANN) for newlinedetecting the attacks in the network. Initially, the big data are collected and newlinepreprocessed using decimal scaling normalization. newline