Design And Implementation Of Emotional Content Classification System For Micro-blog Text

The rapid development of the Internet and information technology has changed people's way of life.A variety of social networking applications,has attracted a large number of Internet users registered to use.Through these network platforms,people can access to news hot spots in time,express their opinions freely.As the current mainstream social network application,micro-blog has a large group of active users,gradually becoming the primary platform for netizens to speak and vent their emotions.A lot of emotional information is flooded with massive micro-blog data.Emotional content classification of these data can not only grasp the trend of the public emotion in time and implement the public opinion monitoring,but also observe the changing state of the emotion,which is beneficial to the management and regulation of the individual emotion.Therefore,it is of great significance in both theoretical research and the application level of the system to accurately identify the user's emotions in the content of the microblogs and to realize the emotional classification of the text content.In the existing research of micro-blog text sentiment analysis,there are some problems such as the imperfect construction of sentiment dictionary and the low accuracy of classification methods.In order to solve these problems,this paper proposes a classification method based on the combination of sentiment dictionary and emotion analysis rules,which can identify the subjective emotion of the microblog users in publishing content.And designs and implements a classification system that can determine and classify the emotional categories of microblogs content.The specific contents of the research are as follows:1.Collected and sorted out the existing well-known Chinese emotion dictionary resources,and constructed the Chinese benchmark emotion dictionary by re-merging.And the emoticons and network words commonly used in the of expression microblogs are collected and emotionally determined to further perfect the micro-blog emotion dictionary.This paper introduces the specific construction rules of each component of the emotional dictionary in detail and uses the multi-point extended SO-PMI algorithm to complete the automatic expansion of the emotional dictionary of the microblogging so as to improve the accuracy rate of the dictionary in recognizing the new emotional word.2.Considering the characteristics of emotion expression in microblog text,this paper analyzes and compares the traditional emotion classification method with the emotion dictionary as the core and the emotion classification method with the machine learning method as the core.Aiming at the shortcomings of the two methods,this paper summarizes the emotion recognition rules of the microblogging,proposes the emotion classification method combining the emotion dictionary with the recognition rule,calculates the values of the various emotion intensities by using the formula,and derives the emotion categories of the microblog text by analyzing the intensity values.Design grouping experiments verify the validity of the classification method,and also prove that the construction of the microblogging emotion dictionary and the proposed emotion recognition rules can effectively improve the accuracy of classification results.3.On the basis of constructing the micro-blog emotion dictionary and the formula for calculating emotional intensity,this paper designs and implements the sentiment classification system for micro-blog text content.The system is divided into four functional modules:data acquisition and management,emotion dictionary construction,text processing and emotional value calculation,emotional category determination.Through the accurate identification and calculation of emotional information in micro-blog text,the emotion classification of the large-scale micro-blog text data set can be automatically classified,and the statistical situation after classification can be displayed.