Identifying DDoS Attacks Using Machine Learning Approaches

Abstract

Cloud computing offers users access to various cloud services, enabling efficient data storage and computational resources with minimal data overhead. However, this convenience comes with a significant risk: Distributed Denial of Service (DDoS) attacks. These attacks leverage multiple compromised computers to target network resources and servers, overwhelming them with messages, malformed packets, and connection requests, ultimately disrupting service for legitimate users. To address this challenge, this project proposes the design of an advanced algorithm that integrates multiple machine-learning techniques. The goal is to develop a model capable of detecting DDoS attacks more accurately. This approach aims to enhance the precision and reliability of DDoS detection, providing a robust defense mechanism for cloud computing environments, analyzing data and its parameters to check any redundancy in data values that may affect prediction results, to remove all the empty or uncertain values and use different types of classifiers to compare the model accuracy of detecting the DDoS attacks and lastly to build a model which aims to provide better model accuracy when compared to that of other models.