Emotion discrimination method and device for minority language text data, equipment and storage medium

Abstract

The invention relates to an emotion discrimination method and device for minority language text data, equipment and a storage medium. The method comprises the following steps: acquiring text data; extracting a plurality of text features of the text data; inputting the text features into a support vector machine model to obtain first evaluation features of the text data; inputting the text data into the mBERT model to obtain a second evaluation feature of the text data; and inputting the first evaluation feature and the second evaluation feature to an emotion discriminator, and outputting an emotion discrimination result of the text data. When the characteristics of the language adopted by the support vector machine model are judged, the emotion judgment accuracy of different languages can be improved, and the generalization ability of model judgment can be well improved by adopting mBERT to learn different language depth semantic vectors to carry out emotion judgment of different languages.

Description

technical field [0001] The present application relates to the field of computer technology, in particular to an emotion discrimination method, device, computer equipment and storage medium for minority language text data. Background technique [0002] With the rapid development and popularization of Internet technology, a large number of users from different countries and languages ​​have made subjective comments on topics such as news and current affairs, policies and regulations, and consumer products on social media such as Twitter and Facebook, which well reflect the individual users. Viewpoints, emotions, attitudes and other important information. Research on how to use text mining technology to understand the emotions expressed by users in various language texts is of great significance to the fields of public opinion monitoring and business analysis. [0003] Sentiment analysis is an important task in the field of natural language processing. The main methods include...

AI Technical Summary

Problems solved by technology

For language-rich languages ​​(such as Chinese, English, etc.), academia and industry have proposed a large number of sentiment analysis methods and tools. However, sentiment analysis tasks for small-language texts still face many challenges, mainly including two aspects: large Most small languages ​​are rich in form, and the expression forms of different language families are quite different, and there is no general method for processing text; there are few related research and public corpus resources with annotations, and it is difficult to support model training

Due to the rich forms of different languages, the differences in the expression forms of different language families, and the uniqueness of text representation, the emotional discrimination effect of low-resource and small languages ​​is unstable.

Images