Research Findings Of Key Users On Corss-Border Social Network

Recently,along with the development of the internet and with the increasingly wide-spread usage of Web 2.0 and MT(Mobile Terminal),conventional channels of transferring information have been becoming more and more internet-oriented,which,in turn,leads to a situation where online social media are now the most popular channel for offering access of information to the public.Having a characteristic of globalization which allows users from all different cultural backgrounds to be able to communicate with each other,social media has been offering a platform on which international news and information are accessed conveniently and rapidly.In recent years,social media has leaded to a diversity of online information instead of purely coming from cultural background of any type,which has caused not only the current ongoing economic reform at the global scale but also several major negative effects on the civilization.Because of being non-boundary and information-sharing,inappropriate information are being spread through social media across the globe rapidly like never before;and it is becoming significantly challenging to monitor quality of the online information given the fact that the amount of information-flow and the rapidity of it are overwhelming.Therefore,it is worth studying and understanding where sources of inappropriate online information are and how to monitor them,which,ultimately,aims at preventing any large-scaled wide-spread of inappropriate online information from happening,locating internet-users who are spreading those information,and mitigating or minimizing the negative effects that would be caused on the society.Considering the relationship between SNS users and historic data of user behaviors,spark will be used as a framework of data processing to design a Spark-oriented methodology that can process the massive data of user behaviors in order to build a weighted social media model.After that,influence of information sharing among users will be built according to PageRank calculation method.Key cross-boundary users are selected based on different boundaries that they are within.Overall,the main tasks of the report are as follows:(1)Since fast-breaking news have a characteristic of being easily wide-spread online,index of ability to share information among internet users will be made by using massive online data posted by the users.After that,the index will be quantified by applying the framework that was mentioned above,followed by transforming those quantified data into the weighted of the social media model.(2)Using the weighted social media model just mentioned above,the weights on the edge between the nodes are the weight contribution of its follower nodes,attention from PageRank algorithm,propose the calculation method of node information dissemination influence based on Spark,thus get each node information transmission ability.(3)By identifying different nodes according to different attributes of user locations,which will increase the sensitivity of fast-breaking news,the report applies the spark-based node identification method and based on Nation attributes in order to match each node to their own boundaries.(4)The report is based on the case analyzing(ASL Ice Bucket Challenge)that analyzed the report's methodology.