Abstract
Internet attacks have sharply increased as a result of growing Internet usage, Cloud Services, the Internet of Things, and the number of linked devices. The most dangerous assault that has a proportionate effect on the target system (a website, server, or other resources connected to a network) is a Distributed Denial-of-Service (DDoS).
In a Distributed Denial-of-Service attack, the target system's services are continuously used by numerous systems, which denies the target system's actual use of those services. Excessive requests to the target result in packet flooding, service breakdowns, wear and tear, and other problems. The result could be a slowdown or even a crash of the target system. The end outcome is that the real users are denied service. The process begins with a master computer pestering weaker computers to attach the target. The number of packets, device activity in the network, statistical analysis of the network, and use of network protocol analyzers can all be used to identify DDoS attacks.
In this project, I plan to visualize the data, create a dashboard for better narrative and understanding, and evaluate the performance of several machine learning algorithms for detecting DDoS attacks. I also intend to use the dataset to apply transfer learning. I will show all of my findings using the Tableau dashboard, which will aid in comprehension.
Network administrators can avoid manual traffic inspection and monitoring by using DDoS attack detection. Additionally, traffic can be diverted and filtered, mitigation plans can be made, and internet browsing is made safer by recognizing the hostile network. DDoS attack detection can, in general, assist us in ensuring network security and the security of internet users worldwide.