ENTC-Crow-ANN: An Efficient Network Traffic Classification using Crow Search based
Artificial Neural Network Algorithm
Abgina Anwar, Institute of Computer Sciences and Information Technology (ICS/IT), The University of Agriculture Peshawar, Pakistan.
Asfandyar Khan, Institute of Computer Sciences and Information Technology (ICS/IT), The University of Agriculture Peshawar, Pakistan.
Noor Ul ArfeenInstitute of Computer Sciences and Information Technology (ICS/IT), The University of Agriculture Peshawar, Pakistan.
Majid Khan, Institute of Computer Sciences and Information Technology (ICS/IT), The University of Agriculture Peshawar, Pakistan.
Corresponding Author:
Noor Ul Arfeen (noorarfeen60@aup.edu.pk)
Abstract:
Due to swift development of technology most of the people are connected with each other through internet for sharing their information and data. Due to huge amount of information and the data shared in network through Internet, the network becomes very congested. It is not clearly defined which networks are busy and free available for sharing the information. Many researchers work to identify network traffic. But all these techniques have some limitations in the term of Accuracy for Network Traffic Classifications. To efficiently classify the network and enhance accuracy, this research proposed a new optimized model which is based on crow search artificial neural network (Crow-ANN). To evaluate the performance of the proposed model, benchmark dataset such as NIMS was taken from UCI Machine Learning repository. In terms of mean square error and the accuracy as compared to the ANN. The result was simulated on 70:30. From simulation result, it shows that the proposed Crow-ANN model have efficiently classified the network traffic. The accuracy of Crow-ANN is 99.9889 with MSE 0.000111473 on 30% testing data which show high performance as compared to the used model such as ANN.
Keywords:
Network Traffic Classification; Machine Learning; Crow Search Algorithm; Artificial Neural Network