A new approach for internet traffic classification: GA-WK-ELM

The classification of internet traffic is one of the very popular topics of the present-day. In particular, the classification studies performed along with the use of machine learning (ML) approaches is increasing a little more with each passing day. In this study, Extreme Learning Machine (ELM) methods were used for the classification of Internet traffic. Kernel Based Extreme Learning Machine (KELM) approach, one of the ELM approaches, was applied to the data. In particular, Genetic Algorithm (GA) based software (GA-WK-ELM) was developed for the selection of the parameters which were used in the (WK-ELM) algorithm in which Wavelet function was used. It was seen that an accuracy rate of over 95% was achieved along with the application developed with GA. The average truth value metric was used in order to compare the performance of the classification performed. In addition, Receiver Operating Characteristic (ROC) curves were also generated.