<![CDATA[Limited network resources and the increasing number of internet users in the current digital era have an impact on high traffic which results in decreased access speed to internet services. This is also a problem that occurs at the Indo Global Mandiri University (UIGM) Palembang, causing access to academic services to be slow. The purpose of this research is to identify the types of network traffic patterns which are then carried out by the process of grouping and visualizing these types of traffic. The data in this study were taken in real-time at the UIGM campus. The data obtained is the result of responses which are then extracted. The extraction results are processed using the Support Vector Machine (SVM) method for the process of grouping and visualizing data. The results of this study can distinguish types of traffic based on communication protocols, namely tcp and udp, where the results of the experiment were carried out six times with the results being the first experiment where 99.7% TCP and 0.1% for UDP, the second experiment 97.6% for TCP and 1.1% for UDP , trial three 99.7 % TCP and 0.2% UDP, trial four 97.5% and 1.3% UDP, trial five 99.5 TCP and 02% UDP, and the sixth or final try 97.7% TCP and 1.1% UDP. The data from the use of the SVM method obtained several types of traffic such as games by 0.4%, mail 0.2%, multimedia 0.4% and the web by 82.8% and this research still produces data that the pattern is not yet recognized by 15.5% Keywords : Network Traffic, Classification, Support Vector Mesin]]>
Copyrights © 0000
