Publication Details

Abstract

DDoS (Distributed Denial of Service) is a major problem in cyber traffic, but it can be easily solved with simple navigation. This is not the case with DDoS (Distributed Denial of Service) systems. This hacker attack exploits buffering and relies on buffer overflow. There is no risk of data loss. Machine learning monitoring models are used to address the problem of data loss. Various machine learning technologies, including Random Forest, K-Nearest Neighbors (KNN), and Logistic Regression, are employed to clarify what is normal and what is not. This studio used the CSE-CICIDS2018, CSE-CICIDS2017, and CICDoS datasets as a test. The dataset is created in different parts of the world, so it can also be used for adding and pruning for testing. The next different search We are trying to classify the DDoS operation using a machine learning classifier. This proposal is used by Other machine learning classifications: Random Forest, KNN, and Logistic Regression. We successfully performed feature scaling with the support of the standard scaler. Slotting is also included in the system. The Random Forest classifier performs better than others. The classification is based on a 97.6% dataset, and KNN corresponds to Logistic Regression, which is based on 97% and 91.1%, respectively. We agree to the use of various technologies such as Supervised Machine Learning, Random Forest, KNN, and Logistic Regression to make the identification test algorithm more efficient. The result is from the Random Forest of other models.