Trace Of Microblogging Topic Initiatior Based On The Influence Of Topic

The popularization of the Internet and network terminal makes virtual social networks increasingly penetrate into every aspect of people's lives,such as the popular online social networks :Sina Weibo,Tencent Weibo,facebook,twitter,etc..Online social networks are more easily joined,higher interactive and open than traditional media,individual behavior can affect other users in these social networks,the information diffuses gradually through the users` relationship and then forms fission mode of spread.However some adverse topics also spread on the open platform and even form public opinion,it seriously endangers public discourse space and social security,and therefore topic initiatior detection research of online social networks can contribute to control tne network public opinion and trace the source of topic to maintain social stability.Existing methods of event initiator detection are mostly sort algorithm based on the similarity between pages as well as time and link relationship,however,this method is not suitable for short text-based microblogging network,we propose a method which trace the source of topic on microblogging network based on the influence of topic,the main work and innovations of this research include the following aspects:Firstly,for the past influence calculating method only concernning on tne single microblog or user,this paper combines the microblogging content with opinion leader to calculate the influence of topic.This method not only can quantitatively measure the topic`s influence within a certain time period,but also find the opinion leader who promote the spread of topic.Secondly,we explore the topic influence trends on the foundation of existing literature for changes of microblogging number in the spread process to provide a basis for the subsequent tracing the source of topic method.Thirdly,we propose a method which can trace the source of topic in microblogging network according to the change of topic influence reverse recursion in the similar topics to search the topic initiator.Also we learn from the classic infectious diseases SIS model in transmission dynamics research which analyse the threshold of information spread widely in order to determine the termination conditions of the recursive process,and ultimately to find the collection of microblogging topic initiator.Finally,we design and implement experiments to verify the effectiveness of the models and methods including obtaining the experimental data,data preprocessing,analysis and discussion of experimental results.The experimental results proved that this method can reduce the range of searching and find the collection of microblogging topic initiator.