Delving Deep into Engagement Prediction of Short Videos
This research is significant for content creators and recommendation systems on social media platforms by providing a method to predict short video engagement from content alone, potentially improving content curation and discovery.
This study addresses the challenge of predicting engagement for newly published short videos by introducing a new dataset of 90,000 Snapchat videos and two novel metrics, normalized average watch percentage (NAWP) and engagement continuation rate (ECR). The research found that traditional video quality assessment scores do not correlate well with engagement, and their method can predict engagement purely from video content.
Understanding and modeling the popularity of User Generated Content (UGC) short videos on social media platforms presents a critical challenge with broad implications for content creators and recommendation systems. This study delves deep into the intricacies of predicting engagement for newly published videos with limited user interactions. Surprisingly, our findings reveal that Mean Opinion Scores from previous video quality assessment datasets do not strongly correlate with video engagement levels. To address this, we introduce a substantial dataset comprising 90,000 real-world UGC short videos from Snapchat. Rather than relying on view count, average watch time, or rate of likes, we propose two metrics: normalized average watch percentage (NAWP) and engagement continuation rate (ECR) to describe the engagement levels of short videos. Comprehensive multi-modal features, including visual content, background music, and text data, are investigated to enhance engagement prediction. With the proposed dataset and two key metrics, our method demonstrates its ability to predict engagements of short videos purely from video content.