SHRSE:Multimodal Short Video Hashtag Recommendation based on Sentiment Enhancement
Anish Gupta, R. W. Blake, Camden Riley, Priya Kapoor · Research Square · 2023
Abstract With the advent of social media platforms the trend of uploading text, pictures, and videos to express personal feelings has resulted in an explosion in the number of short videos on the internet. However, the accumulation of these short videos has resulted in a serious information overload problem, making it challenging for users to find their desired content. Hashtags are an effective solution to alleviate this problem, but users often do not have the habit of adding appropriate hashtags to their tweets. Hence, recommending suitable hashtags has become an important research area in recent years. The multimodality of short videos makes the hashtag recommendation process more challenging than text or image-only recommendation. In this paper, we explore the use of image, audio, and text modalities to achieve a more comprehensive understanding of short videos and design corresponding solutions to improve the accuracy of hashtag recommendation. We provide a comprehensive review of recent research on short video hashtag recommendation and identify the challenges and opportunities for future research.