| Summary: | YouTube, one of the prominent online video-sharing platforms, plays a pivotal role in modern media consumption, making it crucial to understand and predict the view-count dynamics of its videos. The viewership of YouTube videos can be influenced by three distinct sources: subscribers, word-of-mouth, and recommendation systems. This paper presents a comprehensive modelling framework that takes into account the view-count obtained through these three sources, assuming that a single view-count can only be attributed to one of these sources at any given time. We investigate the interplay among these sources in shaping YouTube video view-count dynamics, proposing a novel approach to model and analyse their impact on video popularity. Additionally, the VIKOR multi-criteria decision-making method is employed to validate and rank our proposed models. This study's findings deepen our understanding of the intricate mechanisms within the YouTube ecosystem, offering insights for predicting and managing video viewership.
|
|---|