Understanding Customers' Behaviour through Web Data Mining Models

Abstract

Companies have realized that the customer knowledge contained in web marketing database represent one of the main key to forecast business performance in today's competitive landscape. Appropriate web data mining models are one the best supporting approach to make different marketing decision. Analysing and understanding in advance customers' behaviour can represent the main corporation's strength in planning marketing forecasting. This research want to demonstrate as predictive web data mining models are accurate patterns in predicting marketing performance compared to traditional statistical methods in global business. In addition, particular attention is paid on the identification of the main marketing drivers performed by potential customers before purchasing a given service online. Finally, the criteria based on the loss functions confirm the high predictive power of the web data mining models in detecting the probability of customer conversion.