Listening to the voice of the customers: An early warning system based on sentiment

Abstract

In a global economy, enterprises have to intelligently analyze their data asset in order to stay competitive. With the advent of the Web 2.0, there is a wealth of user generated content which contains valuable information on what customers think about available products and services: the voice of the customers is readily accessible. Listing to and understanding these data can reveal valuable customer insights especially for product quality assessment, improvement and innovation as well as marketing. This report focuses on the use of state-of-the-art text mining techniques for identifying and monitoring the sentiment of customer feedback on Web 2.0 channels over time such as to alert enterprise users to significant increases in negative sentiment as an early indicator of inferior or degrading product quality. © Springer-Verlag Berlin Heidelberg 2013.