Latest News

About CAMQ

Information for Authors

Editorial Board

Browse CAMQ Online

Subscription and Pricing

CAMQ Contacts

CAMQ Home

 

Volume 16,     Number 2,     Summer 2008

 

PROBABILISTIC ESTIMATION AND
PREDICTION OF TELEVISION VIEWER
DEMOGRAPHICS
DAVID BALLANTYNE, AMENDA CHOW,
DAVID DE LA ROSA, ROBERT PICHÉ AND LI XU

Abstract. Advertisers decide which channels and timeslots they will pay to run their ads based on the demographic of the audience. Due to the increase in available channels, the process of accurately assessing the viewing demographic is difficult, particularly for networks with smaller audiences. This means cable television networks receive less ad revenue than traditional networks. Invidi Technologies Inc. proposes that one means of addressing this issue is through personalizing advertising to the current viewing audience via a digital set-top box (DSTB). Two methods are proposed for predicting the current viewing audience based on information obtained by the DSTB: logistic regression and naïve Bayes. The authors demonstrate each approach using a subset of viewer data, and show that the results are similar.

 

Download PDF Files
 
(Subscribers Only)

© 2006-2009 Canadian Applied Mathematics Quarterly (CAMQ)