[Objective] To make up for the vacancy in the method of classification of opinion leaders and evaluate the characteristics of different types of opinion leaders from multiple perspectives. [Methods] A method to classify opinion leaders by community division was proposed, and a two-dimensional analytical framework model was used to comprehensively analyze the influence of opinion leaders from dimensions of network diffusion ability and emotional dominance. The empirical analysis employs Twitter data, and compares the influence of different types of opinion leaders through network analysis and text mining. [Results] Opinion leaders are identified as 3 communities, which rank differently in aspects of network diffusion ability and emotional dominance. The two dimensions show no correlation with an absolute value of correlation coefficient less than 0.3. Compared with the traditional weighted summing method, the two-dimensional matrix analysis can reflect the influence characteristics more comprehensively. [Limitations] In the evaluation of emotional influence of text, only the original text is analyzed, and subsequent studies can make further evaluation by combining the comments. [Conclusions] The proposed methods are helpful to analyze the degree and characteristics of the influence of different types of opinion leaders, helping managers explore the public opinion guidance value of all kinds of opinion leaders in a more targeted way and guide the public opinion direction more effectively in risk events.