《计算机应用研究》|Application Research of Computers

基于模仿创造的网络流行语传播模型及仿真研究

Research on network buzzwords propagation model and its simulation based on imitation and creation

免费全文下载 (已被下载 次)  
获取PDF全文
作者 蒋建洪,李倩倩
机构 桂林电子科技大学 商学院,广西 桂林 541004
统计 摘要被查看 次,已被下载
文章编号 1001-3695(2020)07-005-1940-06
DOI 10.19734/j.issn.1001-3695.2018.11.0950
摘要 随着互联网和智能移动终端的发展,研究网络流行语的传播过程和发展趋势对于网络营销和广告文案的创作具有重要意义。基于SIR传染病模型,综合考虑网民对网络流行语模仿再创造的行为特点,构建新型的网络流行语传播模型,利用神经网络技术结合流行语时序数据对模型进行参数反演,并分别以流行语“佛系”和“确认过眼神”为例进行验证。结果表明,用户的模仿再创造行为是网络流行语传播中后期的主要驱动力;相较SIR模型,该模型着重考虑了网民对流行语的创新行为并运用参数反演方法,其预测准确度更高,模型拟合值与真实数据相比误差更小。进而可以为营销和广告创意人员提供有益的借鉴,并通过预测其发展趋势对舆论进行及时分析和引导。
关键词 SIR传染病模型; 网络流行语传播; 再创造行为; 神经网络; 参数反演
基金项目 国家自然科学基金专项研究资助项目(71940008)
国家教育部人文社科基金资助项目(17YJCZH074)
桂林电子科技大学研究生教育创新计划资助项目(2018YJCX98)
本文URL http://www.arocmag.com/article/01-2020-07-005.html
英文标题 Research on network buzzwords propagation model and its simulation based on imitation and creation
作者英文名 Jiang Jianhong, Li Qianqian
机构英文名 School of Business,Guilin University of Electronic Technology,Guilin Guangxi 541004,China
英文摘要 With the rapid development of the Internet and smart mobile terminals, studying the dissemination process and development trend of online buzzwords are of great significance for the creation of online marketing and advertising copywriting. According to the behavioral characteristics of netizens' imitation and re-creation of network buzzwords, based on SIR epidemic model, this paper proposed a new network buzzwords propagation model. It used neural network technology combined with the timing data of buzzwords to inverse the model parameters, and took the popular words "foxi" and "querenguoyanshen" as exa-mples for verification. The results show that the imitation and re-creation behavior of users is the main driving force in the middle and late stage of network buzzword communication. Compared with the SIR model, network buzzwords propagation model focuses on netizens' innovative behaviors towards buzzwords and applies parameter inversion method, and the accuracy of the model is higher, its fit value is less error than the real data. The research can provide useful reference for marketers and advertisement creatives, and analyze and guide public opinion in time by predicting its development trend.
英文关键词 susceptible-infective-recovered(SIR) epidemic model; network buzzword propagation; re-creation behavior; neural network; parameter inversion
参考文献 查看稿件参考文献
 
收稿日期 2018/11/29
修回日期 2019/3/11
页码 1940-1945
中图分类号 TP391.9
文献标志码 A