<![CDATA[The use of internet traffic in a government office needs to be monitored carefully to obtain efficient and effective use. The internet line that has been provided is an official facility funded by the people's budget, so it needs to be monitored carefully. Domain Name System (DNS) provides rich and interesting data, and can be extracted to reveal information that can be analyzed for various purposes such as security measures, measuring traffic usage levels, bandwidth restrictions, user profiling to other policies implemented in a network. This study aims to make a clustering of the use of internet traffic so as to provide benefits that can be used to Improve Network Services (QoS), make efficiency of bandwidth usage and create user profiles. This research was conducted based on DNS Log which is operated on a network connected to the internet. In this study, it is shown how to consolidate traffic on port 53/udp to collect DNS logs, so that in this way the activities of internet users can be recorded in a centralized server until finally used as a primary data source. The datasets used are information extraction from the DNS Server log file (dnsmasq) which was retrieved for 5 working days in the period of working hours. The total extracted datasets used are 213 records. The available data is then processed to get the target cluster by utilizing the concept of data mining using the K-Mean Clustering method. The results classify the use of internet traffic into 3 clusters, namely high, medium and low. Each cluster consists of Kla1 is 23, Kla2 is 3, and Kla3 is 160. This research can be used as a reference in grouping internet traffic so that communication lines are well and smoothly maintained.]]>
Copyrights © 2022
