Evaluation Model Based on Support Vector Machine for Community Micro-Blog Influence

Abstract

As a widely used medium platform, Micro-blog influence research is a hotspot. The community micro-blog, which is used as an effective tool by social managers in virtual community, has developed rapidly in recent years. As the basis of government micro-blog system in China, the community blog influence has great importance to guide the public popular feelings and guarantee the safety of the virtual social network. The article focuses on the evaluation methods of the community micro-blog influence. Firstly, an index system is presented, both quantitative index and qualitative index are considered based on the mechanism of information dissemination of micro-blog, Then, principal component analysis (PCA) is used to compose these indexes into some comprehensive indexes to simplify the index system, finally, support vector machine (SVM) is adopted for the evaluation model. Practical examples show that the model established in this paper outperforms others in evaluation accuracy.