Social media text processing method and device, equipment and storage medium

Abstract

The invention relates to a social media text processing method and device, equipment and a storage medium. The method comprises the following steps: performing word segmentation processing on a social media text to obtain a plurality of lexical items; for any lexical item, the word vector of the lexical item is determined based on the semantics of the lexical item in the contexts of different application scenes, and the word vector contains the semantics of the lexical item in the contexts of different application scenes, so that the text context feature extraction capability and implicit emotion inference capability can be improved by utilizing the word vector; further, determining a global semantic vector of the social media text based on the respective word vectors of the plurality of lexical items; determining a local semantic vector of the social media text based on the respective word vectors of the plurality of lexical items and the respective weights of the plurality of lexical items in the social media text; and determining the emotion type corresponding to the social media text based on the global semantic vector and the local semantic vector, thereby improving the prediction accuracy of the emotion type of the social media text.

Description

technical field [0001] The present disclosure relates to the field of natural language processing, and in particular to a social media text processing method, device, device and storage medium. Background technique [0002] With the rise of different social networks and social applications, people use social media to share a variety of information on social networks and social applications, such as experience and life, express opinions and opinions, understand customs and cultures around the world, establish and Maintain social relations, market and promote various brand products, and the information it disseminates involves all aspects of modern life. [0003] Sentiment analysis based on social media text has extensive and important application value in different fields, such as social management, business decision-making, and information prediction. At present, deep learning has made breakthroughs in the field of sentiment analysis, mainly through the research directions ...

AI Technical Summary

Problems solved by technology

[0004] However, due to the short length of social media text data, the emergence of new words, and a lot of noise, the data is sparse, which leads to the greatly weakened ability of the neural network to obtain text context; and the social media text data may lack the ability to clearly express emotions. Words (that is, explicit emotional words), and language phenomena that have different polarities in different scenes (for example, in some scenes, the same word expresses positive emotions, while in other scenes, the same word expresses negative emotions), resulting in neural network model When calculating high-order dependent features (including text context features), the benefits are often less than expected, which limits the ability of neural network models to solve difficult problems of implicit emotional inference through contextual features.

Images