Machine Learning based Events Targeting Applications in Online Display Advertising
An important concept in industry and science is the occurrence of rare events: events that occur with low probabilities but have significant impacts. Examples of rare events include the chance of an automobile insurance holder filing a claim or credit card accounts are compromised. The proposed project investigates applications of machine learning algorithms in the analysis and prediction of the occurrence of rare events. Of particular interest in this proposal is the application of rare event targeting to online display advertising that is also the core idea of the flag product of the industrial partner, the Infersystem engine. Outcomes of the research will directly contribute to further improvement on the performance of the Infersystem engine by minimizing the cost and maximizing targeted reach for media campaigns in real time.