Abstract: Consumers are increasingly using social networks to share their opinions about products and services generating large amounts of unstructured Voice of Customer (VoC) data that can be utilized to significantly enhance the knowledge of likely future customer behavior and to analyze their buying intensions. In this paper we propose a methodology that helps organizations to systematically incorporate VoC data into their existing data warehouse. We have chosen the CRISP-DM data mining methodology as the basis of our approach for VoC integration. We focus on VoC data gained mostly from Web 2.0 sources, as this is still an open research problem. The paper proposes a reference model for the analyses of VoC content using a BI (Business Intelligence) process.
Keywords: CRISP-DM, unstructured data, Voice of Customer, Business Intelligence, customer analytics, opinion mining
1 Department of Information Technologies, Faculty of Informatics and Statistics, University of Economics, Prague, Czech Republic, lucie.sperkova@vse.cz
2 Faculty of Engineering and Information Technology, University of Technology, Sydney, Australia, Unicorn College, Prague, Czech Republic, george.feuerlicht@uts.edu.au
[PDF]
Cite: Lucie Sperkova, George Feuerlicht, "Application of CRISP-DM to Voice of Customer and BI integration," Progress in Intelligent Computing and Applications , vol. 5, no. 1, pp. 7-13, March 2016.