JSM 2012 Home

JSM 2012 Online Program

The views expressed here are those of the individual authors and not necessarily those of the JSM sponsors, their officers, or their staff.

Online Program Home

Abstract Details

Activity Number: 240
Type: Contributed
Date/Time: Monday, July 30, 2012 : 2:00 PM to 3:50 PM
Sponsor: Section on Statistics in Marketing
Abstract - #305357
Title: Hierarchical Bayesian Classifier for High-Dimensional Online Behavior Data Applied to Consumer Segment Prediction and Behavior Insights
Author(s): Amit Phansalkar*+
Companies: Compete Inc.
Address: 43 Bennington St, Newton, MA, 02458, United States
Keywords: Bayesian ; online behavioral modeling ; consumer segmentation ; high dimensional data ; targeted marketing
Abstract:

Consumer segment prediction based on high dimensional online click-stream behavior data poses challenges due to the large volumes of granular data with low signal to noise ratio. Accurate predictions of users into segments aids in understanding consumer behaviors and effectively marketing products to the appropriate consumers. Compete has successfully amassed online behavioral data from over 2 million users. Our objective is to classify and predict shopper segments for the entire user panel based on sample of behavioral and survey data to be used for model training. We have developed an approach using hierarchical Bayesian classification to identify the behavioral features that can accurately predict user segments using training data. When considering large set of behavioral features with a skewed distribution across these features traditional approaches result in very low accuracy rates due to the noise. We will present accuracy rate comparison results with traditional methods. A case study showing application of our methods to the online panelist behavioral data will be presented along with accuracy results and behavior insights for consumer segments.


The address information is for the authors that have a + after their name.
Authors who are presenting talks have a * after their name.

Back to the full JSM 2012 program




2012 JSM Online Program Home

For information, contact jsm@amstat.org or phone (888) 231-3473.

If you have questions about the Continuing Education program, please contact the Education Department.