Stochastic Strategies for Addressable Advertising

Recently, the nature of media advertising has changed dramatically due to the introduction of highly efficient addressable advertising models. The latter have the ability to deliver different ads during one advertising spot to the viewers of same network, depending on household demographic profiles. The new models provide industry with the opportunity to maximize advertising revenue and profitability and in return allow for more broad accessibility of advanced technologies. Essential to these techniques is the estimation of a capability to fulfill advertising orders within the new bandwidth constraints of the system. One particular problem for delivering multiple advertisements is the allocation of resources to different networks and the prediction of collisions between commercial breaks throughout the set of networks. The objective of this proposal is to construct optimal stochastic, statistical/probabilistic models for ad break interleaving and the correlations between ad breaks, TV viewing, TV programs, viewer demographics, and other aspects.

Faculty Supervisor:

Ivan Mizera

Student:

Partner:

Invidi Technologies Corporation

Discipline:

Mathematics

Sector:

Information and cultural industries

University:

University of Alberta

Program: