Data sentiment analysis method, device, electronic equipment and medium

Abstract

The present invention provides a data sentiment analysis method, device, electronic equipment, and medium, including: a method, including: collecting comment information; obtaining user ratings in the comment information; and obtaining the emotional level of the comment information based on a dictionary using keyword matching and dictionary rules The emotional polarity includes neutral, negative, and positive; analyze the probability of comment information belonging to different emotional polarities based on machine learning; use the mapping method to convert the score, emotional polarity, and emotional polarity into The same range; the score, emotional polarity and probability of emotional polarity converted to the same range are combined by weighted voting to obtain the emotional polarity and score to which the comment information belongs. The invention has good generalization effect and high analysis accuracy.

Description

technical field [0001] The present invention relates to the technical field of data mining, and more specifically, to a data sentiment analysis method, device, electronic equipment, and computer-readable storage medium. Background technique [0002] Sentiment analysis technology currently mainly adopts methods based on sentiment dictionary, method based on machine learning technology or method based on deep learning. [0003] Based on the sentiment dictionary method, the text is preprocessed by word segmentation and stop word processing, and then the text is matched with strings using the constructed sentiment dictionary to mine positive and negative information. Dictionary-based text matching algorithms are relatively simple. Traverse the words in the word-segmented sentence one by one, and if the word hits the dictionary, the corresponding weight will be processed. [0004] The sentiment classification based on dictionary sentiment is simple and easy to implement, and it...

AI Technical Summary

Problems solved by technology

But there are still many deficiencies: (1) the precision is not high

Language is a highly complex thing, using simple linear superposition will obviously cause a great loss of precision

Word weights are also not static and difficult to be accurate

(2) Difficult to find new words

(3) Dictionary construction is difficult

This problem brings great difficulty to word segmentation and word vector

Images