Analysis of Anomaly with Machine Learning Based Model for Detecting HTTP DDoS Attack

At present, almost every device is connected to the internet for communication. Information also can be quickly obtained via the Internet. The Internet offers various helpful services like news portals, video streaming, social media, e-commerce, and more. Thus, service providers must prioritize ensuring the availability of the services. One of the threats that service providers face is distributed denial of service (DDoS) attacks. DDoS attacks happen when someone sends lots of requests to a website's server, making it unusable for others who need to use it. Most of the information on the Internet is transmitted primarily via the Hypertext Transfer Protocol (HTTP). It serves as one of the pathways for hackers to execute DDoS attacks. Detecting anomalies in HTTP attacks forms a vital part of cybersecurity. This involves identifying anomalous patterns that do not conform to expected behavior, potentially indicative of a variety of attacks. Machine learning techniques are increasingly used in this field due to their ability to learn and adapt to new patterns. This research represents the performance analysis of several machine learning techniques, such as Logistic Regression, k-Nearest Neighbors, Decision Tree, Naïve Bayes, and Random Forest. The experimental result shows the highest accuracy was obtained using the Random Forest algorithm with an accuracy value of 99.998% on a dataset that had been oversampled and feature selection applied.