Decision support system for bandwidth expansion according to the increment of users

Abstract

Considering the local context, in Sri Lanka internet speediness has become a strain issue domestically and in organization status making the promising promises of internet providers of delivering the fastest internet speed, an abortive attempt. One of the major reasons to that is giving a fixed bandwidth to the area, or fixed number of EnodeB s (EUTRAN Node B / Evolved Node B in 4g LTE). But with the increase of usage that given bandwidth becomes insufficient causing a lot of network traffic, mostly in peak times it almost becomes unusable as same as with EnodeBs becomes insufficient because too many clients connecting to the same node. The thorough analysis about this particular circumstance leads to understand that this issue has been occurred in different countries and most of the expertise has looked upon the matter. With the understanding that bandwidth expansion according to the user increment is a quite common concern which has been undergone with several suggested solutions; this research is going to explore the matter in terms of the local context considering the sample of consumer usage data of Lankabell (pvt) Ltd. The collected data will be processed through data mining algorithms. The objectives of this paper are to identify the internet speed decrements, high-risk area and low-risk area customers by using the data mining techniques like clustering.