Emotion classification method based on support vector machine

Abstract

The invention discloses an emotion classification method based on a support vector machine. The emotion classification method comprises an emotion classification system, wherein the emotion classification system comprises a crawler module CPM, a feature word and training sample generating module TGM, an SVM classification module and a visualization module VM which are in communication connection in sequence, wherein the feature word and training sample generating module TGM comprises a feature word extraction module and a training sample selection module; the emotion visualization method by adopting the emotion visualization system comprises the following steps: A, the crawler module CPM performs data acquisition from a station; B, the feature word and training sample generating module TGM selects the feature words and the training samples; C, SVM classification module performs classifying; and D, the visualization module VM shows an analysis result at a Web end. By adoption of the emotion visualization method based on the support vector machine, the comment information, published on microblog and forums, of a user can be accurately classified according to the emotion of the user so as to find out conditions of public sentiments.

Description

technical field [0001] The invention relates to a support vector machine and public opinion analysis technology, in particular to an emotion classification method based on a support vector machine. Background technique [0002] With the rapid development of the Internet, the data on the Internet has shown explosive growth. According to incomplete statistics, within one minute, 100,000 new microblogs were added on Twitter. In China, Sina Weibo has 650 million users, with 46 million daily active users, Tencent Weibo has 620 million users, and about 100 million daily active users; not only that, the valuable information in traditional forum websites is about 1 year old. About 100 million. Behind such a large number of active users and the rich and emotional comments they post, there is a lot of valuable information hidden. The analysis of this information can help to discover commenters’ emotions towards specific subjects, for example: microblog / forum users’ comments on the ...

AI Technical Summary

Problems solved by technology

"If you don't consider the scene of the text and only judge the emotion of the sentence itself, you will often get inconsistent judgment results.

Images