Improved Machine Learning Techniques for Precise DoS Attack Forecasting in Cloud Security

Abstract

 One of the fundamental motives of Cloud based computing for the use of technologies of current era that based on Internet. The concept of cloud computing has exploded in popularity, and the reason for this is the cost-effective transmission, storage, and intensive computation that it offers. The goal is to provide end-users with remote storage and data analysis capabilities utilising shared computer resources, lowering an individual's overall cost. Consumers, on the other hand, are still hesitant to use this technology owing to security and privacy concerns. This paper provides a thorough overview of the different risks and technological security problems associated with cloud computing. We use the UNSW dataset to train the supervised machine learning models. We then test these models with ISOT dataset. The algorithm's accuracy for DoS and probe attacks was investigated, and the findings were given as confusion matrices. Cloud computing has changed the technological scope by offering cost-effective transmission, storage, and computation. It’s security especially on Distributed Denial of Service Attacks remains a major concern. This study uses two datasets, UNSW and ISOT, to train and test supervised machine learning models for the prediction of DoS attacks. The model used achieved a remarkable accuracy of 99.6%. These findings present the ability of machine learning to improve cloud security in the near term.We have achieved an accuracy of 99.6% to predict a DoS attack. We present our results and argue that more research in the field of machine learning is still required for its applicability to the cloud security.