Abstract

This paper introduces how Sapphire varies from the standard conventional methodologies used for ranking YouTube Videos. By implementing Natural Language Processing, specifically Computer Linguistics, Sapphire enhances YouTube Video ranking by specifically weighing videos based on how much content they covers rather than their Statistics. The conventional Watch Time algorithm ranks videos based on features such as viewership status, likes, subscriptions, etc [Fyf20]. This Whitepaper delves into what makes Sapphire distinct and domain specific from the Watch Time Algorithm.