Method and device for identifying emotional polarity of microblog short text and storage medium

Abstract

The invention provides a method and device for identifying emotional polarity of microblog short text and a storage medium. The method comprises the following steps: extracting emoticon, repeated punctuation marks and the number of English capital words in a microblog text as microblog context sentiment characteristics tc; calculating a prior polarity characteristic value SenScore (t) of each microblog T; generating a fixed length vector y'of the microblog T by using a deep convolutional neural network; And determining the sentiment polarity of the microblog short text based on the microblog context sentiment feature tc, the prior polarity feature value SenScore (t) and the fixed length vector y '. According to the method, the problems of dimension explosion and data sparsity of microblogvector representation are solved, emotional feature vector representation of different levels is combined, emotional semantic features of microblog texts can be described more comprehensively and variously, and therefore the emotional polarity recognition performance of the microblog texts is improved.

Description

technical field [0001] The invention relates to the technical field of natural language processing, in particular to a method, a device and a storage medium for recognizing the emotional polarity of short microblog texts. Background technique [0002] Emerging social media represented by Weibo has become an important medium for Internet users to obtain news information, social interaction, self-expression, share opinions, disseminate information and social participation. main platform. As of June 2018, the global microblogging service site Twitter has 326 million monthly active users, the Chinese microblogging platform Sina Weibo has 431 million monthly active users, and 190 million daily active users. Hundreds of millions of active users come from different social and cultural backgrounds and spread all over the world, generating a large amount of information containing users' subjective opinions and emotions every moment. These text messages contain various emotions, emo...

AI Technical Summary

Problems solved by technology

[0003] Microblog text is short, grammatical structure is incomplete, and the expression is random and full of noise, making the emotional polarity recognition of microblog text face greater challenges

Existing classification research mainly uses artificial methods to construct classification features, and the bag-of-words model is often used for classification features, which is difficult to describe word order information and faces the problems of dimensionality disaster and data sparseness, which makes the judgment of the emotional tendency of Weibo texts slow, and is not suitable for microblog texts. quick analysis of the situation

In addition, microblog texts often have "explicit praise and implicit derogation" or implicit expressions, and it is difficult to discover and describe such implicit semantic relationships by artificially constructing features.

Images