摘要
[目的/意义]准确预测舆情观点的演化态势,对化解舆情危机具有重要意义。[方法/过程]针对HK模型用于用户观点演化预测存在的不足,进行如下改进:首先,从和当前事件相似度高于阈值的历史博文中获取用户的初始观点值;其次,利用用户间交互行为以及关注关系计算亲密度,修订用户信任阈值;最后,融合用户全局影响力、用户间交互度以及观点接近度计算用户间的综合信任度,突出用户间观点影响权重的差异性。最终利用改进的HK模型实现对社交用户观点演化预测。[结果/结论]实验结果表明,改进的HK模型用于观点演化预测具有较低的MSE值和MAE值,可为舆情发展提供有效预测。
[Purpose/Significance]Accurate prediction of the evolution of public opinion is significance in resolving public opinion crises.[Method/Process]To solve the shortcomings of the existing HK model for predicting the evolution of social users'opinions,the following improvement were given:Firstly,the initial opinion values of users were obtained from historical blog posts with similarity to the current event above a threshold.Secondly,the user confidence threshold was revised by calculating the affinity using the interaction behavior between users and the following relationship.Then,the global influence of users,the interaction between users and the proximity of opinions were combined to calculate the comprehensive trust degree between users,highlighting the differences in the weight of opinion influence between users.Finally,the improved HK model was used to predict the evolution of social users'opinion.[Result/Conclusion]The experimental results show that the improved HK model for opinion evolution prediction has lower MSE and MAE values and can provide effective predictions for opinion development.
作者
吴树芳
郑依静
朱杰
Wu Shufang;Zheng Yijing;Zhu Jie(School of Management,Hebei University,Baoding 071000,China;College of Mathematics and Information Science,Hebei University,Baoding 071000,China)
出处
《现代情报》
北大核心
2024年第2期142-151,共10页
Journal of Modern Information
基金
河北省人文社会科学研究重大课题攻关项目“基于大数据的河北省网络治理机制研究”(项目编号:ZD202102)
河北大学社会科学研究重大课题培育项目“网络生态驱动下谣言传播及阻断机制研究”(项目编号:2021HPY004)
河北省自然科学基金项目“基于语义结构的自适应图卷积跨模态特征学习方法研究”(项目编号:F2022511001)。
关键词
HK模型
观点演化预测
社交用户
HK model
opinion evolution prediction
social users