Enhancing the Detection of DDoS Attacks in Cloud using Linear Discriminant Algorithm

The environment of Cloud Computing (CC) offer, its users a platform to exchange resources, services, and information. With an increased demand for data, the cloud faces security threats and vulnerabilities. Organizations are more vulnerable to Distributed Denial of Service (DDoS) attacks and hackers are focusing on the application layer. Due to low resource consumption, it is hard to detect such attacks caused by low and slow DDoS. Slow read HTTP DDoS is one among them, it is a tricky attack that interrupts regular network activities in the application layer through abnormal traffic. Linear Discriminant Algorithm (LDA)is a conventional technique for detecting attacks in the cloud. Random Forest, Naive Bayes, and Decision tree classifiers are applied for comparison. The proposed algorithm shows a high accuracy rate of 98.7% for detecting DDoS attacks in the cloud.