Intelligents systems with classification of traffic information in corporate computer networks

The classification of data on the internet in order to make internet use more efficient has an important place especially for network administrators managing corporate networks. Studies for the classification of internet traffic have increased recently. By these studies, it is aimed to increase the quality of service on the network, use the network efficiently, create the service packages and offer them to the users. The first classification method used for the classification of the internet traffic was the classification for the use of port numbers. This classification method has already lost its validity although it was an effective and quick method of classification for the first usage times of the internet. Another classification method used for the classification of network traffic is called as payload-based classification or deep packet inspection. This approach is based on the principle of classification by identifying signatures on packets flowing on the network. Another method of classification of the internet traffic which is commonly used in our day and has been also selected for this study is the extreme learning machine based approaches. It is based on classifying by the use of more statistical methods and algorithms collecting the flow information on the network. For the classification of the internet traffic, extreme learning machines (ELM), which were hardly ever used in previous studies, were used. In order to compare the performance of ELM, support vector machines (SVM), Naive Bayes (NB) and artificial neural networks (ANN) from learning machine algorithms used before were also compared by applying to data set. It was observed that a faster and higher level performance was achieved in the classification made with ELM algorithms compared to other learning machine algorithms. Classification was made by applying both classical and kernel based ELM (KELM) approaches to data. The receiver operating characteristic (ROC) curves were created for each class of the classifiers. In particular, wavelet function which uses the KELM algorithm used parameters for the selection of a good degree of genetic algorithm (GA) based software (GA-WF-KELM) was developed.