数据与计算发展前沿 ›› 2026, Vol. 8 ›› Issue (1): 183-194.

CSTR: 32002.14.jfdc.CN10-1649/TP.2026.01.015

doi: 10.11871/jfdc.issn.2096-742X.2026.01.015

• 技术与应用 • 上一篇    下一篇

基于多模态特征的短视频热度预测研究——以抖音平台为例

米赛雪(),张琪,张士豪*(),李根   

  1. 中国人民公安大学，信息网络安全学院，北京 100038
  • 收稿日期:2025-06-05 出版日期:2026-02-20 发布日期:2026-02-02
  • 通讯作者: 张士豪
  • 作者简介:米赛雪，中国人民公安大学，硕士研究生，主要研究方向为社交媒体流行度预测与网络舆情分析。
    本文中主要负责数据建模、实验设计与论文的撰写与修改。
    MI Saixue is a master’s student at the People’s Public Security University of China. Her research interests include social media popularity prediction and online public opinion analysis.
    In this paper, she is mainly responsible for data modeling, experimental design, as well as manuscript writing and revision.
    E-mail: 1026081425@qq.com|张士豪，中国人民公安大学，硕士，讲师，主要研究方向为信息隐藏、社会网络分析等。本文主要承担工作为论文内容修改。
    ZHANG Shihao, holding a master, s degree, is a lecturer at the People’s Public Security University of China. His research interests include data hiding and social network analysis.
    In this paper, he is mainly responsible for revising the manuscript.
    E-mail: zhangshihao@ppsuc.edu.cn
  • 基金资助:
    中央高校基本科研业务费(2024JKF02ZK09)

Research on Short Video Popularity Prediction Based on Multimodal Features: A Case Study of Douyin Platform

MI Saixue(),ZHANG Qi,ZHANG Shihao*(),LI Gen   

  1. School of Information and Cyber Security, People’s Public Security University of China, Beijing 100038, China
  • Received:2025-06-05 Online:2026-02-20 Published:2026-02-02
  • Contact: ZHANG Shihao

RichHTML

1

PDF

0

摘要：

【目的】短视频已成为网络舆情传播的重要载体，准确预测短视频热度对内容监管和舆情分析至关重要。然而，现有研究在特征提取和时序建模方面存在以下不足：一是特征维度单一，未能充分利用多模态数据；二是传统线性方法难以刻画短视频“冷启动-爆发-衰减”的热度变化规律。为此，本研究提出一种基于多模态特征的短视频热度预测方法。【方法】首先，构建多模态特征体系，涵盖用户影响力、作者影响力、音视频质量及内容特征、评论特征及热度特征。其次，采用随机森林模型进行非线性建模，以捕捉特征间的复杂关联，并提高预测视频热度能力。【结果】实验表明，所提方法在短视频热度预测任务中表现优异，F1分数达69.3%，较基线模型提升13.7个百分点。AUC值达到71.3%，较基线模型提升了16个百分点。【结论】基于多模态特征的热度预测方法能显著提升短视频热度预测的准确性，为网络舆情分析与内容管理提供有效技术支持。

关键词: 短视频, 热度预测, 多模态特征, 用户影响力, 随机森林

Abstract:

[Objective] Short videos have become a crucial medium for online public opinion dissemination, making accurate popularity prediction vital for content moderation and public sentiment analysis. However, existing studies exhibit limitations in feature extraction and temporal modeling: First, the unidimensional feature analysis fails to fully leverage multimodal data sources. Second, conventional linear approaches prove inadequate in characterizing the nonlinear popularity dynamics of short videos, particularly the distinctive “cold-start-explosion-decay” lifecycle patterns. To address these gaps, this study proposes a multimodal feature-based approach for short video popularity prediction. [Methods] First, a multidimensional feature system is constructed, encompassing user influence, author influence, audiovisual quality and content features, comment features, and interaction features. Second, the Random Forest model is employed for nonlinear modeling to capture complex feature interactions and improve the ability to predict video heat. [Results] Experimental results demonstrate the superior performance of the proposed method in short video popularity prediction tasks, achieving an F1-score of 69.3%, representing a 13.7 percentage point improvement over the baseline model. The AUC value reaches 71.3%, showing a 16 percentage point enhancement compared to the baseline. [Conclusions] The multimodal feature-based approach significantly improves prediction accuracy, offering a robust technical solution for online public opinion analysis and content governance..

Key words: short video, popularity prediction, multimodal features, user influence, random forest