Automated Denial of Service Detection Using Moth Flame Optimization With Machine Learning in Cloud Environment

Abstract

Denial of Service (DoS) attack detection refers to preventing and detecting malicious attempts to make network resources or services unavailable to its intended users. DoS attacks is been a main concern for organizations since they can disturb the accessibility of critical services and cause economic losses. In cloud environments, the mitigation and detection of such attacks were very challenging because of the dynamic and distributed nature of infrastructure. In this regard, Machine Learning (ML) methods are potential in identifying DoS attacks, by using network traffic features to find unusual anomalies or paradigms that may specify an attack. This research work introduces an automated Denial of Service Detection using Moth Flame optimization with Machine Learning (DoSD-MFOML) technique in cloud environment. The DoSD-MFOML technique recognizes DoS attacks and the MFO algorithm is used for feature selection purposes to attain improved results. The detection of DoS attacks takes place using extreme gradient boosting (XGBoost) classifier. Finally, the DoSD-MFOML technique employs grey wolf optimizer (GWO) algorithm for the parameter tuning procedure. The performance validation of the DoSD-MFOML method is tested on benchmark dataset and the outcomes are studied under several measures. The experimental outcome confirms the increased performances of the DoSD-MFOML technique for DoS attack detection purposes.