融合情感特征和可解释性的弹幕视频传播效果预测模型  

A Prediction Model for Danmaku Video's Propagation Effects with Sentiment Features and Interpretability

作　　者：倪渊[1,2] 华君鹏 张健 杨翠芬[1,2] 张腾 Ni Yuan;Hua Junpeng;Zhang Jian;Yang Cuifen;Zhang Teng(School of Economics&Management,Beijing Information Science&Technology University,Beijing 100192,China;Beijing Key Laboratory of Big Data Decision Making for Green Development,Beijing 100192,China)

机构地区：[1]北京信息科技大学经济管理学院,北京100192 [2]绿色发展大数据决策北京市重点实验室,北京100192

出　　处：《数据分析与知识发现》2025年第2期146-158,共13页Data Analysis and Knowledge Discovery

基　　金：国家重点研发计划青年科学家项目(项目编号:2021YFF0900200)的研究成果之一。

摘　　要：【目的】在弹幕视频传播效果预测模型中融入情感特征以提升预测效果,利用模型可解释性量化各特征变量的影响。【方法】基于BERT-BILSTM对弹幕视频传播影响因素情感特征进行提取。提出基于PCACVRFE-RF-XGBoost的组合预测模型对弹幕视频的传播效果进行预测,基于1 515部文化弹幕视频的传播数据进行实证分析。【结果】挖掘出31个变量覆盖了信息质量、信源可信性和信息传播感知质量三方面特征。在弹幕情感特征提取准确率上,BERT-BILSTM模型在测试集中积极和消极分类的精确率分别达到0.81和0.85,F1值达到0.84。实验结果表明,基于CVRFE-RF-XGBoost构建的弹幕视频传播效果预测结果优于SVR、BP神经网络模型。【局限】弹幕文本情感分析的粒度仍待细化。【结论】所提模型为情感特征复杂、高动态性的弹幕视频传播效果预测提供新方法。通过样本实证结果表明,信源可信度的特征贡献度高于信息质量,这意味着信源可信度对弹幕视频传播效果的影响程度更深,其中,媒介平台口碑、媒介平台专业性、个人影响力、内容发布频次等特征尤为关键。[Objective]This paper integrates sentiment features into the prediction model for danmaku video propagation effects to improve prediction performance and to quantify the impact of various feature variables using model interpretability.[Methods]We extracted sentiment features influencing danmaku video propagation using the BERT-BiLSTM model.Then,we proposed a combined prediction model based on PCA-CVRFE-RFXGBoost to predict the propagation effect of danmaku videos.Finally,we empirically analyzed using propagation data from 1,515 cultural danmaku videos.[Results]Thirty-one variables were identified,covering three aspects:information quality,source credibility,and perceived quality of information dissemination.For sentiment feature extraction,the BERT-BiLSTM model achieved 0.81 and 0.85 precision rates for positive and negative classifications in the test set,with an F1 score of 0.84.Our prediction model based on CRFE-RFR-XGBoost showed an improvement across four evaluation metrics compared to SVM and BP neural network models.[Limitations]The granularity of sentiment analysis for danmaku text requires further refinement.[Conclusions]The proposed model provides a novel approach for predicting the propagation effects of danmaku videos with complex and highly dynamic sentiment features.Empirical results show that source credibility contributes more to propagation effects than information quality.Key features include media platform reputation,media platform professionalism,personal influence,and content publishing frequency.

关 键 词：传播效果预测 情感特征 XGBoost 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]