Rumor Blocking Considering User Experience And Outbreak Threshold In Online Social Network

With the rapid development of Web 2.0 and mobile Internet technologies,social media,such as Sina Weibo and WeChat,have become the main platforms for users to access information,spread information and communicate with friends.However,the popularity and openness of online social network(OSN),unfortunately,turns out to be a double-edged sword,which not only provides people with convenient channels of information interaction,but also reduces the costs of spreading misinformation and then become the breeding ground for rumors.For example,the“Red,Yellow and Blue Kindergarten”in in 2017 and the“Shuanghuanglian”in 2020,etc.Moreover,the characteristics of immediacy,openness,interactivity and convenience speed up and expand the diffusion of rumors,which can easily result in public opinion crisis and greatly test capabilities of social governance and crisis response of government agencies.However,in the face of modern social media,the traditional measures have little effect on controlling rumor propagation.Therefore,the deep disclosure for the spread and evolution mechanism of rumors on social medias under the background of public emergencies with the effective strategy for the control of rumors is not only the cyberspace governance problem to be solved by the administrators of the network platforms and the related departments of the government,but also one of the key scientific problems to be studied in the fields of information system and emergency management.Therefore,based on the law of rumor propagation in OSN,this paper constructs a dynamic rumor propagation model,and from the perspective of public opinion crisis,considers user experience and outbreak threshold constraints,and designs two rumor blocking strategies,that is RBUE and RBOTUE algorithms,proves and analyses their performance by simulation experiment.In the research of rumor spreading models,based on SIR model,a dynamic rumor spreading model,called PISIR model,is constructed considering both the overall popularity of rumors at macro-level and the individual spreading tendency of rumors at micro-level based on Ising model.Moreover,an energy model is applied to describe the changing process of rumor attraction,and calculate the influence of node on node by the relative influence weight when constructing rumor spreading model.To reveal the spreading laws of rumors in OSN,the rumors related to the Novel Coronavirus Pneumonia Epidemic(COVID-19)since the end of 2019 are given as an example.We grabbed the rumors about the COVID-19 on Sina Weibo,and obtained the evolution process of rumor spreading through data preprocessing and smoothing.A multi-peak gaussian distribution is introduced to simulate the overall popularity of rumors,which conforms to the actual situation of the multi-peak characteristic of rumor spreading in OSN.In the research of rumor blocking,combined with individual characteristics of online users and structure characteristics of OSN,a hyperbolic discount utility function is introduced to construct the user experience model.Taking the user experience model as constraint,two rumor blocking algorithms are proposed,called Static and Dynamic RBUE algorithms respectively.The goal of algorithms is to minimize the spread range of rumors,under the constraint of the user experience model,select and block specific nodes to maximize the reduction of rumors influence range in OSN.The experimental results show that Static RBUE and Dynamic RBUE algorithms have better rumor blocking performance in NW network,the earlier the blocking start time is,the larger the proportion of blocking nodes is,and the smaller the rumor spreading range is.They have different adaptability under different situations.Dynamic RBUE algorithm can achieve better rumor blocking performance when the blocking starts earlier,while Static RBUE algorithm can achieve better rumor blocking performance when the blocking starts later.Furthermore,from the perspective of public opinion crisis,the outbreak threshold and user experience of rumor spreading are introduced as constraints,two rumor blocking algorithms are proposed,called 1-Hop and 2-Hop RBOTUE algorithms respectively.The goal of algorithms is to minimize the cost of blocking rumors,under the constraints,the unspecific blocked nodes are selected and blocked,so that the rumor spreading range is always lower than the rumor outbreak warning line,which can restrain the rumor outbreak and avoid public opinion crisis,and then achieve the balance between the cost and the effect of rumor restraining.In addition,this paper also proves the effectiveness of the strategies based on theoretical analysis,and then provides theoretical basis for the government and relevant departments to formulate scientific and effective rumor governance measures.Experimental results show both 1-Hop and 2-Hop RBOTUE algorithms can achieve lower rumor infection rate and require fewer blocked nodes under the same conditions compared with other algorithms,which means that the proposed algorithms have better blocking performance with lower restraining cost of rumors in the mainstream social networks,and the two algorithms also have different adaptability to different OSNs.In order to keep the rumor infection rate lower than the outbreak threshold,the earlier the detection start time is and the longer the blocking duration is,the less the number of blocked nodes is,that is,the lower the restraining cost of rumors is and the better the blocking performance of rumors is.However,in Sina Weibo network,when the number of blocked nodes is large,appropriate blocking duration should be selected to ensure user experience and reduce the number of blocked nodes.Moreover,in NW network,compared with Static and Dynamic RBUE algorithms,1-hop and 2-hop RBOTUE algorithms have lower infection rate and better blocking performance.The RBOTUE algorithms all have better blocking performance to different overall popularity of rumors in NW network.