A Micro-blog Public Opinion Evolution Analysis Method Based On LDA

As its convenience,micro-blog has been the general public's favorite,more and more people are willing to express their views and emotions in the micro-blog,which makes it become an important place for public opinion and its dissemination.Timely grasp the evolutionary trend of public opinion is import to Monitoring System.The main contents of this paper are divided into two parts.The first part is micro-blog emotional polarity analysis and emotional value calculation,micro-blog will be divided into micro-blog expression and micro-blog text.Emotional expressions were constructed in emotional value computation,we use inverted index type to calculate Emotions value.In the micro-blog text emotion value calculation,the paper first analyzes the factors that affect the emotional value of the text,and then uses the emotion dictionary and semantic context rules synthetically,and puts forward a micro-blog emotion value calculation method.On the basis of the LDA model,the second part introduces the concept of thematic heredity and proposes an online topic evolution model based on LDA and Sentiment Evolution.In the analysis of public opinion evolution using S-OLDA,we not only consider the change of topic content and topic intensity over time,but also change the subjective emotion over time as an indicator of public opinion evolution.The emotional value of the topic is calculated using the micro-blog emotional value calculation method presented in the first part.The micro-blog emotional value calculation method proposed in this paper has a good effect in micro-blogging emotional polarity classification.Compared with OLDA,S-OLDA model has lower model perplexity and has better effect in micro-blog topic mining.