Natural Adsorbents for Effective Treatment of Dye Wastewater from Textile Industries

Abstract

The growing reliance of IT industries on Software-defined networks to meet networking needs has put the security considerations in these networks at the forefront. Cyber security advancements over the years have begun to incorporate artificial intelligence/machine learning techniques to build new and improved security measures to counter the IT-related threats that loom in the world today. In the proposed chapter we have described the cutting-edge innovations and related development of Intrusion detection systems (IDS) for Software-defined networks (SDNs) using Machine Learning (ML)/Deep Learning (DL) techniques. The chapter discusses the methods that ML/DL researchers have proposed to design IDS. This includes ML techniques like Support Vector Machines, Naïve-Bayes, Decision trees, Random forests, Logistic Regression, etc., and deep learning techniques like Neural networks, RNN, LSTM, and CNN. Apart from the aforementioned techniques, the chapter also incorporates the research work wherein deep generative models like Generative adversarial networks (GANs) and Variational autoencoder (VAE) have been used for designing IDS for SDNs. We have presented the results obtained in the published results in terms of precision, accuracy, recall, and F1-score and have found the anomaly-based IDS to be more effective than the signature-based IDS.