A Prediction and Classification Process for DDoS Attacks Using Machine Learning
Mangesh D. Salunke, Vijay U. Rathod, Yogesh Kisan Mali, Ravindra S. Tambe, Anup A. Dange, Suhas Ramdas Kothavle
Abstract
The popularity of computer networks in our daily life has grown recently as a result of the extensive use of the Internet. Due to server flaws, hackers can access computers using not only well-known attack types but also newer, more advanced, and harder to spot ones. One of the maximum popular defenses is the Intrusion Detection System (IDS), which uses machine learning techniques to train on a pre-collected dataset to defend computers from them. In this paper, the authors examine the present practice of Distributed Denial of Service (DDoS) attacks. DDoS attacks make use of restrictions that are applicable to any arrangement asset, such as the website's framework for an authorized organization. However, it is crucial to use the most recent dataset in order to identify the most recent DDoS activities. The authors employed an old-fashioned KDD dataset. In this research, a machine learning technique was used to identify and predict the sorts of DDoS attacks. The K Nearest Neighbor, Random Forest, and Decision Tree are three machine-learning-based IDSs approaches that we propose in this paper.