Improving the performance of network traffic prediction for academic organization by using association rule mining

Network traffic prediction for academic organizations is essential to managing and selecting the best routing path. Since overload traffic is a major problem that delays data transmission in network system and causes some data loss, this research demonstrates an approach to predict network traffic on data transmission in the network system by using association rule discovery which is one of the data mining techniques. Durations of semester and vacation were employed to support prediction. In addition, amount of incoming and outgoing data volume, durations and sizes of routing path were applied to create a model of network traffic prediction on data transmission in network system. The results of this study pointed out that association rule discovery could predict network traffic during semester and vacation for the following day. This could be useful factor for network routing selection and could improve highest performance of data transmission in network system.