Efficient Binary Classifier for DoS Detection using Quine-McCluskey

The proposed research considers the difficulty of efficiently detecting Denial of Service (DoS) attacks in real-time networks, concentrating on the complexity of computation and accuracy of detection. The central point is the hard task of dealing with large sets of features and the redundancy of the irrelevant data. This leads to the situation that the classification is not only not accurate but also not delivered in time. To overthrow this, the system uses the **Quine-McCluskey optimization technique** to choose the most suitable characteristics, so that the redundancy is eliminated and at the same time the classification process is improved. This way it is guaranteed that the binary classifier will be able to recognize new intrusion patterns and improve decision accuracy. The result of the study is an improved detection system that can be scaled down and offers cybersecurity that is based on feature dependency and decision consistency, hence, it is very reliable in the case of different network environments. It is this way that we can equip the system to prevent real-time DoS attacks more effectively and also let it be more responsive. The approach aids in the development of reliable and responsive systems to withstand DoS attacks in real-time scenarios.