| The objective of this research is to build a predictive model that can predict customer behavior, based on other attributes of the customer. The data used for this research belongs to Company ABC, a financial services company. Based on the behavior predicted, customers would then be segmented into different groups. Each group would then be approached with a customized marketing strategy, to increase business. Several statistical techniques like correlation and principal component analysis were employed to first analyze the data, reduce the data set and build predictive models. A comparison was carried out between Clustering and Neural network models, to determine the most suitable model for the current scenario. It was observed that in spite of high percentage error, clustering is more favorable than neural network, due to its approach in segmenting data into various groups. (Abstract shortened by UMI.)... |
