Public opinion information emotion classification method

Abstract

The invention discloses a public opinion information emotion classification method. The method comprises the following steps: collecting a certain amount of public opinion information data and carrying out emotion labeling; collecting the public opinion information data needing to be predicted; training a Google BERT model by using the marked training set; utilizing a Google BERT model to classify emotion types of the unlabeled public opinion information data; and visually displaying the public opinion information processing result through echarts, wherein the emotion classification accuracy of the method on a test set can reach 82%, and the processing result can be flexibly and visually displayed in a visual mode.

Description

technical field [0001] The invention belongs to the field of natural language processing, and in particular relates to a method for classifying emotions of public opinion information, which is used for classifying the emotions of public opinion information and visualizing processing results. Background technique [0002] With the rapid development of mobile Internet and social software, more and more people began to express their opinions and comments on various social software. An important component of information has important reference value: it can reflect the public's views on certain events, and can represent the response of some policies or measures to a certain extent. Therefore, if you want to understand public opinion on certain events, or the response to policies or measures, these public opinion information from social software play an important role. [0003] At present, the technology of collecting various types of information on the Internet, that is, web cr...

AI Technical Summary

Problems solved by technology

However, these models still have the problem of insufficient classification accuracy (the classification accuracy of TextRCNN on the test set is about 65%), which also directly affects the accuracy of public opinion information data analysis, so efforts need to be made to improve the accuracy of emotion classification

[0005] In addition, it is not enough to comprehensively, accurately and intuitively reflect the real public opinion information only by classifying the sentiment of the public opinion information data and displaying the results, because the public opinion information data has many attributes, including the time of generation, the number of reposts, the number of comments, and the number of likes. Only by combining these data that reflect key information can we better reflect the popularity information and changing trends of real public opinion information

Images