Multi-modal and Metadata Capture Model for Micro Video Popularity Prediction

TL;DR

The paper tackles micro video popularity prediction (MVPP) by proposing the 3M model, a multi-modal framework that integrates video, audio, descriptions, and metadata. It combines a retriever-based memory bank with cross-modal attention and a semi-supervised metadata module, then fuses outputs through an integrated network to improve predictive accuracy. Key findings show the author-enhanced ensemble (E) outperforms baselines on validation across multiple targets (HEART, SHARE, COMMENT, PLAY) using $MSE$ and $PLCC$ metrics, demonstrating the value of incorporating diverse modalities and incomplete data handling. The work has practical implications for content optimization, advertising strategy, and market opportunity identification in short-video platforms.

Abstract

As short videos have become the primary form of content consumption across various industries, accurately predicting their popularity has become key to enhancing user engagement and optimizing business strategies. This report presents a solution for the 2024 INFORMS Data Mining Challenge, focusing on our developed 3M model (Multi-modal and Metadata Capture Model), which is a multi-modal popularity prediction model. The 3M model integrates video, audio, descriptions, and metadata to fully explore the multidimensional information of short videos. We employ a retriever-based method to retrieve relevant instances from a multi-modal memory bank, filtering similar videos based on visual, acoustic, and text-based features for prediction. Additionally, we apply a random masking method combined with a semi-supervised model for incomplete multi-modalities to leverage the metadata of videos. Ultimately, we use a network to synthesize both approaches, significantly improving the accuracy of predictions. Compared to traditional tag-based algorithms, our model outperforms existing methods on the validation set, showing a notable increase in prediction accuracy. Our research not only offers a new perspective on understanding the drivers of short video popularity but also provides valuable data support for identifying market opportunities, optimizing advertising strategies, and enhancing content creation. We believe that the innovative methodology proposed in this report provides practical tools and valuable insights for professionals in the field of short video popularity prediction, helping them effectively address future challenges.

Figures (2)

• Figure 1: multi-modal information extraction module
• Figure 2: metadata capture module