| As the deepen of integration of Internet and traditional industry in recent years,microblogging,a typical social networking applications,performs its advantages in interpersonal communication and information diffusion,which draws the increasing attention of enterprises.And now,microblogging has become important channels of information dissemination and customer management tools for many companies.Microblogging's content is simple,easy to release and transmit.And its topic can be detected in the mass of information,which has significant implications for enterprises to monitor public opinion,analyze competitive intelligence and social marketing of their products.Furthermore,users' feedback of enterprise microblogging can be found in a variety of topics by analyzing of each topic.User's feedback can improve the dissemination effect of enterprise microblogging.However,massive enterprise microblogging data not only provides opportunities for enterprise information extraction,but also poses a great challenge to the enterprise.After study of the current related researches on enterprise microblogging,we find that the existing LDA model could not only find out the potential topic,but also an effective solution to the word polysemy.However,it ignores the interaction between microblogging and fans,which causing the enterprises lose some valuable information.Thus,this paper selects the accounts of Club of Huawei as the data source.Combining with LDA model,the super-network research method was used to construct the super-network model of the microblogging topic of the enterprise.After the deep analysis of the enterprise microblogging communication focus,commentary hotspot,communication core and its theme,a set of practical methods for the discovery are suggested to help enterprise find valuable themes,analyze the user interests and improve the corporate microblogging publishing strategy to better attract fans and increase the value of corporate microblogging. |
PDF Full Text Request