A Machine Learning-Based Classification and Prediction Technique for DDoS Attacks

Aim:

          We proposed a complete systematic approach to detect DDOS attack using machine learning algorithm.

 Abstract:

          Distributed network attacks are referred to as Distributed Denial of Service (DDoS)attacks. These attacks take advantage of specific limitations that apply to any arrangement asset, such as the framework of the authorized organization’s site. In the existing research study. It is necessary to work with the latest dataset to identify the current state of DDoS attacks. In this presented work, used a machine learning approach to predict DDoS attack types.For this purpose, used Random Forest and XGBoost classification algorithms. To access the research proposed a complete framework for DDoS attacks prediction. To meet the proposed objective, we used UNWS-np-15 dataset and Python was used as a simulator. After applying the machine learning models, we generated a confusion matrix for identification of the model performance.In the first classification, the results showed that both Precision (PR) and Recall (RE) are 88% for the Random Forest algorithm. In the second classification, the results showed that both precision(PR) and Recall(RE) are approximately 90% for the XGBoost algorithm

Synopsis:

      Distributed network attacks are referred to, usually,as Distributed Denial of Service (DDoS) attack. A DDoS attack sends different requests (with IP spoofing) to the target web assets to exceed the site’s ability to handle various requests, at a given time,and make the site unable to operate effectively and efficiently_ even for the legitimate users of the network. Typically,the target of various DDoS attacks are web applications and business websites; and the attacker may have different goals.

Existing System:

         CNN and RNN both are two different algorithms that can be used for different purposes. For example, CNN is used for feature extraction and RNN is used for regression in time series data utilization. Though both CNN and RNN based model producing accurate results, it is very long and time consuming process.

Problem Definition:

       The authors used the CNN and RNN model for intrusion detection.This is a very long and time-consuming process. Therefore, it is very important to perform advanced machine learning techniques to model optimization that train the best model for highly accurate work.

Proposed System:

        Among the machine learning techniques, random forest and XGBoost both are powerful supervised learning models.Both are applicable and used for classification problems. The random forest algorithm is approximately 100 times faster than other algorithms and best working for classification problems.

 Advantage:

            It is approximately 100 times faster than the random forest and best for forbid data analysis. Both are simple and faster than other algorithm in terms of execution times.