Comparative Study of IDS against DDOS Attack using Machine Learning Algorithms

Data storage, bandwidth, and processing speed are all growing in the IT industry. The resulting increase in cyberattacks and threats necessitates the use of cutting -edge technology to employ a creative and predictive security approach. Using the information available, a model will be built by analysing various data sets and monitoring trends. (DDoS) are one of the prevalent problems and assaults wreaking havoc on internet-connected computer equipment? This study compares the working of several ML-based classifiers for detecting DDoS attacks before they occur. This experiment analysed data from the benchmark KDD-Cup-1999 DDoS attack. I have three distinct types of selection techniques to select essential features in the context of DDoS detection. The results show that feature selection methods can help domain experts understand the intrusion system's main hidden patterns and features during DDoS detection. The main challenges they faced are that DDOS attacks are more complex and attacks are coming from many places which makes them difficult to detect. The proposed sample learns to recognize regular network traffic to detect ICMP, TCP, and UDP DDoS traffic as it arrives. Tests show that machine learning algorithms can correctly classify traffic into regular and DDoS. The finding has lasting implications for many industries, including defence, financial institutions, and healthcare, and because other businesses need sophisticated intrusion detection techniques. The future direction of the proposed method is to implement the model in different datasets.