An Integral Sentiment Intelligent Classification Method for Complex Review Texts Based on Synthetic Deep Capsule Networks

Abstract

In order to alleviate the influence of word spelling mistakes, wrong order and mixed emotions on the final grasp of the overall sentiment score of reviews, these noise and mixed emotions can be dealt with in a targeted and orderly manner. The present invention designs a comprehensive deep capsule network classification model , simulating the logical steps of human reading, by capturing feature information at the word level, phrase level, and sentence level, respectively, to model reviews, specifically, modeling at the word level and phrase level with misspelling errors and word order errors Corresponding to the noise problem, the modeling at the sentence level corresponds to the emotional mixing problem, that is, each short sentence is regarded as a synonymy, and the impact of different synonymy groups on the final overall emotional attitude is dynamically considered. In terms of implementation, the BERT WordPiece vector and convolution can be used as word-level and phrase-level features, and then the capsule network can be used to obtain the final vector representation at the sentence level for classification.

Description

technical field [0001] The invention relates to the technical field of artificial intelligence, in particular to an overall emotional intelligence classification method for complex review texts based on a comprehensive deep capsule network. Background technique [0002] As online shopping and consumption have become one of the mainstream lifestyles in today's society, commenting on various products and services on the Internet has become a convenient and effective way for people to publish their satisfaction with products and services, and the amount of user comment data has exploded. growth. How to grasp the attitudes and intentions of users expressed in these review texts plays a vital role in the improvement, marketing and promotion of products and services. However, the comment texts filled in by these users often contain spelling mistakes (or abbreviations), word order errors and other noises and mixed expressions of positive and negative emotions and attitudes, which ...

AI Technical Summary

Problems solved by technology

[0007] The present invention proposes an overall emotional intelligent classification method for complex review texts based on a comprehensive deep capsule network, which is used to solve or at least partially solve the technical problem of poor classification effect existing in the methods in the prior art

Images