Correspondence: Alan Abrahams, Virginia Polytechnic Institute and State University 1007 Pamplin Hall Virginia Tech Blacksburg, VA 24061-0235 USA. Tel: +1 540 231 5887; e-mail: abra@vt.edu

¹received his bachelor's degree in Computer Science (2003) and master's degree in computer applications (2006) from the University of Mumbai. Currently, he is an MS candidate in Computer and Information Science (Engineering) at the University of Pennsylvania. He has worked as a Teaching Assistant for 'Project Management' at the University of Pennsylvania (Fall 2006). His research interest focuses on Artificial intelligence, Machine learning, Data mining and analysis.

²received his bachelors degree in Systems Engineering from the University of Pennsylvania's Engineering School and his bachelors of Science in Economics in 2006 from the Wharton School at the University of Pennsylvania. His concentrations at Wharton were in Operations and Information Management and Finance. He joined The Boston Consulting Group in New York in 2007. His research interests include Machine learning, Data mining and Entrepreneurial finance.

³is an assistant professor at Virginia Polytechnic Institute and State University. He has previously taught in the Department of Operations and Information Management at The Wharton School, University of Pennsylvania, and in the Department of Informatics at the University of Pretoria. He holds a PhD in Computer Science from the University of Cambridge, and a Bachelor of Business Science with honours in information systems from the University of Cape Town. His primary research area is entrepreneurial decision support, including identification and valuation of new technology applications, societal welfare initiatives (HIV/AIDS clinical decision support), automated contract management, and simulations for modelling knowledge growth and transfer.

Received 23 July 2007; Revised 23 July 2007.

Abstract

We demonstrate the use of decision tree induction, employing both C4.5 and Profit Optimal (SBP) algorithms, to discover profitable locations for a young startup firm to sell their product, pet insurance. We use publicly available data including US Census data and veterinary surgery location data as our data sources and use the potential profits generated by each of the algorithms as key performance metrics. We show how our findings link to general business behaviour and performance, by describing the implications of our findings for marketing strategy at the pet insurance company.