|
Activity Number:
|
396
|
|
Type:
|
Invited
|
|
Date/Time:
|
Wednesday, August 5, 2009 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Statistics and Marketing
|
| Abstract - #303038 |
|
Title:
|
Modeling Competition Amongst Web Sites Using Clickstream Data
|
|
Author(s):
|
Alan Montgomery*+
|
|
Companies:
|
Carnegie Mellon University
|
|
Address:
|
Tepper School of Business, Posner Hall 255A, Pittsburgh, PA, 15213,
|
|
Keywords:
|
Bayesian Models ; Choice Modeling ; Web browsing
|
|
Abstract:
|
Web browsers constantly make choices about which web site to view and whether to continue browsing at the current web site or switch to an alternative one. In this research we propose a choice model to predict user browsing behavior that implicitly defines competition. This model is formulated in a Bayesian hierarchical model and leverages information from user choices as well as the information content contained within the text of the pages visited. In our modeling framework, we use text mining techniques to categorize topics that are viewed and combine this with time series models that capture user switching behavior across topics. We find that users tend to view related topics together and that most web browsers are not persistent in seeking out new information.
|
