An intrusion identification and prevention for cloud computing: From the perspective of deep learning

Abstract

As the Internet industry has connected the globe using advancements in computer network from LAN to Cloud Infrastructure, Fashions and Developments in Internet of Things (IoT), Cloud Infrastructures, E-Commerce, Banking and Healthcare are allowing unimaginable access to devastating attacks.Intrusion detection is a critical component of information security, and the essential technique is the ability to properly recognise diverse network assaults. In this article, the methodology of an intrusion detection scheme utilising deep learning technique which is based on Fuzzy Min Max Neural Networks-Based Intrusion Detection System (FMMNN-IDS) has been proposed. The fuzzy min-max learning algorithm, an expansion-contraction method that can learn nonlinear class boundaries in a single pass through the data and gives the capacity to incorporate new and improve current classes without retraining, is used to identify the min-max points.Furthermore, the article investigated the model's performance in terms of binary and multiclass classification, as well as the count of learning rate and neurons affect the model's performance. The performance of FMMNN-IDS is compared with state-of-art approaches determined in the literatures. The FMMNN-IDS model increases accuracy rate of intrusion detection, achieves better intrusion detection rate and reduced error rate.