Visual Analytics In Social Media Marketing With WeChat Official Account Recommendation

With the rapid development of social media platforms,emerging social media platforms such as WeChat and Weibo have caused a profound transformation in the traditional marketing field.In the WeChat platform,social media marketing relies on the WeChat public account and the friend circle social network to push soft advertising,which is to use WeChat articles as a carrier to promote products and services.In the real marketing process,it is hard for marketers to find WeChat public accounts that meet the characteristics of products and services from a large number of WeChat public accounts,which bring significant challenges to marketers.This paper proposes a data visualization system based on text data mining,which is designed to help marketers to analyze massive data of WeChat public accounts,discover proper public accounts for marketing tasks,improve WeChat marketing efficiency,and complete WeChat marketing tasks.Firstly,at the WeChat public account level,this paper mainly focuses on the WeChat public account data,and proposes an influencing model of the WeChat public account to evaluate the influence of WeChat public account in the industry.Term frequency and inverse document frequency are applied to calculate the similarity of the WeChat public accounts.Clustering the WeChat public accounts to analyze the relationship between them,and the clustering result is projected into the visualization view.Secondly,at the semantics analysis level of WeChat articles,this paper uses the text topic model LDA(Latent Dirichlet Allocation)to discover the topic distribution of the industry and present the marketing dynamics on the WeChat platform.In order to facilitate users to quickly and accurately understand the topic of the WeChat article,this paper extract the article abstract by using methods of generating articles abstracts automatically.The word distributed representation is used to generate the word vector of article words for analyzing the semantic relationship of WeChat article keywords.Finally,this paper designs an interactive data visualization system to present the characteristics of WeChat article data,help marketers accomplish marketing tasks,and validate our system through case studies.