Text sentiment analysis method and system based on reinforcement learning

Abstract

The invention relates to the technical field of text sentiment analysis, and discloses a text sentiment analysis method based on reinforcement learning, which comprises the following steps of: obtaining a text to be subjected to sentiment analysis, and performing word segmentation processing on the text to be subjected to sentiment analysis by utilizing a bidirectional maximum matching method to obtain a word segmentation text of the text to be subjected to sentiment analysis; calculating the association strength between words in the word segmentation text by utilizing the dependency graph; according to the association strength between the words, iteratively calculating importance scores of the words by utilizing a TextRank algorithm so as to obtain keywords in the word segmentation text;converting the word segmentation text keywords into word vectors by using a Huffman tree; and performing sentiment classification on the word vectors by utilizing a sentiment classification model based on reinforcement learning, and adding sentiment factors into the sentiment classification model to optimize the model. The invention further provides a text sentiment analysis system based on reinforcement learning. According to the invention, sentiment analysis of the text is realized.

Description

technical field [0001] The invention relates to the technical field of text sentiment analysis, in particular to a text sentiment analysis method and system based on reinforcement learning. Background technique [0002] With the development of Internet technology and the improvement of people's living standards, the Internet has entered thousands of households, and more and more users like to express their views and opinions online. Nowadays, there are a large number of short texts with strong subjectivity and refined words published by users on various Internet platforms such as e-commerce platforms, social networking sites, Weibo, and Twitter. These views and attitudes are helping businesses grasp market conditions and assist consumers. For consumption decision-making and corporate crisis public relations, how to complete sentiment analysis of short texts has become a hot topic in current research. [0003] In the existing text sentiment analysis technology, there are mai...

AI Technical Summary

Problems solved by technology

[0003] In the existing text sentiment analysis technology, there are mainly sentiment analysis methods based on sentiment lexicon and machine learning. Among them, the sentiment lexicon-based method first needs to create a sentiment lexicon, and then according to the emotional tendency of words or phrases in the lexicon and Intensity information, to realize the sentiment classification of text, the advantage of this method is that it does not need to label the data set, but it also has some shortcomings, such as poor scalability, not only depends on the model and quality of the dictionary, but also depends on the rules In addition, the construction of the emotional dictionary is very complicated, time-consuming and labor-intensive, and the matching process requires complete and accurate matching; while the sentiment analysis method based on machine learning has problems such as the difficulty of effective feature selection and the lack of labeled training corpus

Images