A video monetization system and method are disclosed to run an auction to maximize the revenue from a sequence of multiple video ads, while minimizing the impact on the user's watch time. The system uses two predictive models, implemented with a machine learning algorithm that also takes into account user behavior/history. The method uses whole-commercial session features like the number of ads shown, the length of each ad, and the sequence of ads for selecting optimal commercial breaks to display to a user. Online video platforms may use full-session auction to optimize the entire sequence of video ads that they show within commercial breaks, in order to maximize revenue effectively.

Creative Commons License

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.