To search, Click below search items.


All Published Papers Search Service


Predicting required bandwidth for educational institutes using prediction techniques in data mining (Case Study: Qom Payame Noor University)


Mohammad Niknam Pirzadeh, Behrouz Minaei, Mohammad Tari, Jafar Pour Amini


Vol. 11  No. 6  pp. 260-266


In this paper, we are going to analyze the information of a universitys database using prediction techniques in data mining in order to predict internet bandwidth required for future periods. This prediction can help the university apply policies to utilize the existing facilities in the optimal way. Further, considering the high cost of bandwidth, by making a guess at peak hours of traffic in network, the university can formulate a special strategy to purchase or lease the internet bandwidth. In this paper, using Clementine software, we have dealt with application of prediction algorithms to Payame Noor Universitys database that after the stages of preparation and cleanup has included sixty successive months of the maximum bandwidth used by the university. Finally, we concluded that using these techniques it is possible to predict the bandwidth required for the next month of the university with 94.4 percent accuracy.


data mining, algorithm, bandwidth, classification, neural network, prediction