Research On Weibo Rumor Recognition Model

The emergence of mobile Internet technologies and platforms has provided new models and channels for social interaction.People can use the mobile devices to surf the Internet anytime and anywhere.They can get information consultation at the first time and communicate with colleagues and friends without leaving home.In recent years,with the rapid spread of social platforms such as WeChat and Weibo,people are increasingly relying on real-time sharing of information through social platforms.The emergence of social platforms has facilitated information sharing and obtaining.However,it has also become a breeding ground for rumor information.Compared with traditional word-of-mouth,network rumors with social platforms spread faster and wider,and the damage caused may be far greater than traditional methods of communication.Therefore,in order to avoid the serious harm caused by widespread rumors,how to identify rumor information in a timely and effective manner and adopt effective control strategies to block its dissemination is a research hotspot in the field of network information security.At present,in order to solve the problem of rumors,social application providers such as WeChat and Weibo have launched various rumors or official accounts to publish official information timely with real-time monitoring and user reporting as the main measures.They block the spread of rumors by delete posts and many other ways.On the other hand,researchers also have made many efforts to identify rumors.They use common classification models to transform identification problems into classification problems.By combining shallow and deep features that clearly distinguish between rumors and non-rumors,they construct a classification model with extracted features and classification algorithms.Some scholars also use the analogy method to establish a rumor recognition model based on existing models or theories.However,the existing research results still have certain limitations and the accuracy of the recognition results still has room for improvement.Therefore,this article aims at the microblog rumors and takes the rumor recognition as the object,mainly completing the following work.Firstly,collecting and sorting out various types of literature in the field of rumors,we understand the main methods used in current research and their limitations.Defining the concepts of rumors in this study and clarify the scope of research object through sorting out the conceptual evolution of rumors.Secondly,we extract the basic characteristics of user,communication and content identified by rumors.Based on the shallow features and the trust factors,a trust-based microblog rumor recognition model is constructed.Thirdly,the process of generating and transmitting rumors from the process of generation and propagation to explosion is analogous to the detonation and explosion process of explosives in explosive mechanics.The energy parameters of microblog information and the characteristic parameters of rumors are combined to construct the microblog rumor information explosion identification model.Finally,combined with the research content of this paper,from the perspective of process and subject,the corresponding rumor regulation strategy is put forward,and the research work of this paper is summarized.