Predicting Customer Preference of Mobile Service using Neural Network

Abstract

In countries where several Mobile Communication Service providers operate, it is imperative on the service providers to recognize the aspects of their services that will attract new customers in order to help them stay on top of the competition. This is known to be expensive in terms of money and time. To address this problem, we present a Feed-forward Back-propagation Neural Network (FBNN) that is aimed at learning potential customers’ “would-be” pattern of choosing a Mobile Service based on selected criteria. The Neural Network is tested on sample data and predictions made to the same effect as already mentioned. The results show that the Neural Network is adequate for predicting customer preference of a mobile service. This research concentrates on predicting new customer preferences as opposed to the popular notion of models predicting (existing) customer churn.

Keywords: Feed-forward Back-propagation, Mobile Service Provider, Neural Network, Prediction.