In an application-layer distributed denial of service (App-DDoS) attack, zombie computers bring down the victim server with valid requests. Intrusion detection systems (IDS) cannot identify these requests since they have legal forms of standard TCP connections. Researchers have suggested several techniques for detecting App-DDoS traffic. There is, however, no clear distinction between legitimate and attack traffic. In this paper, we go a step further and propose a Machine Learning (ML) solution by combining the Radial Basis Function (RBF) neural network with the cuckoo search algorithm to detect App-DDoS traffic. We begin by collecting training data and cleaning them, then applying data normalizing and finding an optimal subset of features using the Genetic Algorithm (GA). Next, an RBF neural network is trained by the optimal subset of features and the optimizer algorithm of cuckoo search. Finally, we compare our proposed technique to the well-known k-nearest neighbor (k-NN), Bootstrap Aggregation (Bagging), Support Vector Machine (SVM), Multi-layer Perceptron) MLP, and (Recurrent Neural Network) RNN methods. Our technique outperforms previous standard and well-known ML techniques as it has the lowest error rate according to error metrics. Moreover, according to standard performance metrics, the results of the experiments demonstrate that our proposed technique detects App-DDoS traffic more accurately than previous techniques.

Distributed Denial of Service (DDoS) attacks are a growing threat to online services, and various methods have been developed to detect them. However, past research has mainly focused on identifying attack patterns and types, without specifically addressing the role of freely available DDoS attack tools in the escalation of these attacks. This study aims to fill this gap by investigating the impact of the easy availability of DDoS attack tools on the frequency and severity of attacks. In this paper, a machine learning solution to detect DDoS attacks is proposed, which employs a feature selection technique to enhance its speed and efficiency, resulting in a substantial reduction in the feature subset. The provided evaluation metrics demonstrate that the model has a high accuracy level of 99.9%, a precision score of 96%, a recall score of 98%, and an F1 score of 97%. Moreover, the examination found that by utilizing a deliberate approach for feature selection, our model's efficacy was massively raised.