“…Among the reviewed solutions, researches such as [6], [7], [8], [15], [19], and [14], are solutions that implicate high complexity and it is not possible to include simple mechanisms that would prevent the analysis of all the data in real time in order to detect a DoS attack, meaning that it is difficult to reduce the DoS attack detection time. In case of the solutions in [9], [10], [11], [20], and [22], they applied machine learning algorithms that were trained with normal datasets and attack datasets. In those cases, instead of analyzing all real time data with the trained models, those solutions can apply methods to only consider suspicious data in the analysis with the trained models, reducing detection time without affecting the efficiency.…”