Sentence sentiment polarity analysis method and device and storage medium

Abstract

The invention provides a sentence sentiment polarity analysis method and device and a computer readable storage medium. The sentence sentiment polarity analysis method comprises the steps that sentiment polarity of a sentence is analyzed through a machine learning model, a result vector representing the sentiment polarity of the sentence is obtained, and the machine learning model comprises at least two of a decision tree model, a classifier model, a convolutional neural network model and a recurrent neural network model; and fitting the result vector by using a logistic regression model to obtain the sentiment polarity of the sentence. In the embodiment of the invention, deep semantic analysis is carried out through a sentence polarity analysis method of multi-model deep fusion, and the polarity judgment accuracy is high.

Description

Technical field [0001] The present invention relates to the field of information technology, and in particular to a method, device and computer-readable storage medium for analyzing sentiment polarity of sentences. Background technique [0002] Public opinion analysis is a process of in-depth thinking processing, analysis and research on the public opinion on this issue according to the needs of a specific issue, and obtaining relevant conclusions. In the analysis of public opinion, people usually care about the emotions expressed in some news or microblogs, and judge the user's emotional tendency toward the product through emotions, while the sentiment judgment of the longer text is based on the sentiment polarity judgment of the sentence. Therefore, how to accurately judge the sentiment tendency of a sentence is an important problem to be solved urgently. [0003] The prior art sentiment polarity analysis methods are usually based on simple matching of sentiment dictionaries, an...

AI Technical Summary

Problems solved by technology

[0003] The sentiment polarity analysis method in the prior art is usually based on the simple matching of the sentiment dictionary, and the judgment error rate for neutral sentences is too high

In addition, the existing technology is usually based on simple matching and weight accumulation based on the sentiment dictionary, coupled with some simple machine learning models, such as the fasttext (fast text classifier) ​​model, which is inaccurate for judging the sentiment of sentences with deep semantics , unable to meet the task requirements

Furthermore, the prior art has insufficient accuracy for judging the polarity of longer sentences

Images