Research on The Text Emotion of Multinomial Naïve Bayes Integration Algorithm

With the development of the Internet industry, the number of Internet users grows explosively every year. In the Internet era, public opinion can play a decisive role in the process of society. Therefore, the emotional analysis of netizens' comments is particularly important. In this paper, Multinomial Naive Bayes method, which is widely used in word frequency analysis, is used to generate several weak classifiers, and then AdaBoost method is used to obtain a strong classifier by linear combination of several weak classifiers. Experiments show that the combination algorithm of Multinomial Naive Bayes and AdaBoost can classify text emotion more accurately.