A system and method are disclosed that rank videos on a website of a service provider during a user search. The method includes an algorithm for ranking the content based on criteria such as session length, video watch time within a session, video completion and number of unique videos uploaded by the same creator. The algorithm determines if a video has an increased or a decreased ranking based on an aggregate of the criteria. The method involves calculating each of these elements to generate a total score across a cluster of sessions or a specific session, for each video. The system then presents the videos to the user based on their aggregate ranking. The disclosed method reduces server-side CPU cycles and network traffic, as well as client-side battery expenditure and network traffic, by efficiently identifying which content a user will be satisfied with watching.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Justin Lewis and Davies, Scott, "Method Of Ranking Videos During A Search", Technical Disclosure Commons, (April 24, 2017)