Social Commerce Opinion Leader Recognition Based On User Activity And Emotion In The Comments

With the rapid rise of social media and E-commerce,social commerce has developed rapidly.Social commerce mainly refers to an E-commerce model that uses social media and online media to facilitate the purchase and sale of goods through social interaction and user-generated content.In the social commerce community,users mainly share their original content to communicate.In this process,influential users will gradually evolve into opinion leaders in the community,which will have a greater impact on other users in product selection and purchase,and guide the community's topic discussion.Therefore,it's import to idenfy the correct opinion leaders in social commerce for both brand parties and the platforms.Based on the review of related literatures at home and abroad,this paper selects a typical third-party social commerce platform—the Xiaohongshu APP as a specific research object.Combined with social network analysis and sentiment analysis techniques,a recognition model of opinion leaders in social commerce communities is proposed.This paper firstly pre-sorts the user's activity from the perspective of network structure,and filters the target user set of opinion leaders.This stage draws on the basic idea of PageRank algorithm,comprehensively considers the user's behavior characteristics,and designs the ActivityRank algorithm based on multi-index comprehensive evaluation method.Secondly,from the perspective of emotion in comments,the emotional influence analysis of the opinion leaders concentrated in the target users is carried out by analyzing the emotional tendency of user comments.In this stage,according to the application background of social commerce,an emotional dictionary including four kinds of dictionary of basic,domain,modifier and negative words is constructed.The emotional polarity phrase is used as the basic unit of emotional polarity calculation,and the layers are progressively derived to obtain the emotional influence of the user.Finally,the network structure and comment emotions are integrated to construct the final opinion leader recognition model.In order to make the research have a specific application background and real data support,this paper uses the Xiaohongshu APP community as an example to verify the accuracy and efficiency of the model to identify social commerce opinion leaders.Through the recognition model,the obtained opinion leaders are at the core of the network structure,and reflect the influence of opinion leaders on other users emotionally,which makes the identification of opinion leaders more scientific and reasonable.In a word,this paper comprehensively considers the network structure of social commerce community and emotions in comments,and constructs a social commerce opinion leader recognition model.While improving the efficiency of the algorithm,it also overcomes the dilemma that the sentiment analysis method is not suitable for large-scale social network analysis,so that the evaluation is better in line with the objective reality and improves the accuracy of opinion leader recognition.