Text sentiment analysis method based on new word extension and complex sentence pattern extension

Abstract

The invention discloses a text sentiment analysis method based on new word extension and complex sentence pattern extension. The method comprises the following steps: S1, firstly, constructing a basicdictionary according to an existing sentiment dictionary, and cleaning and screening the existing dictionary; s2, performing data cleaning on the imported Chinese corpus, and expanding sentiment words in a specific field according to a basic sentiment dictionary; s3, on the basis of the existing method, finding new words in the specific field and adding the new words into the basic dictionary bycombining word frequency, part-of-speech and similarity calculation; s4, analyzing the structure of the Chinese sentence pattern, summarizing and concluding a sentence pattern model, and judging the sentiment polarity of the sentence through different models; and S5, obtaining an algorithm selector suitable for the method, and integrating the dictionary and the sentence pattern model to obtain a sentence polarity result. Compared with a traditional emotion dictionary and machine learning method, the method focuses on short text sentence emotion recognition in the specific field, and the accuracy and recall rate are obviously improved.

Description

technical field [0001] The invention belongs to the field of text classification sentiment analysis, in particular to an analysis method for short text sentiment classification in a specific field. Background technique [0002] The convenience of interaction has made the Internet one of the main ways for people to express their opinions and communicate with each other more and more. Subjective texts generated on the Internet contain a lot of useful emotional information. More and more people are accustomed to expressing their positive, neutral or negative emotions, as well as their preferences for using products on these platforms. Therefore, reviews on various shopping websites, Weibo, forums and other platforms will become the basis for consumers to make purchase decisions. [0003] Due to the huge amount of network evaluation information, it is infeasible to rely on manual methods, the efficiency is very low, and it is difficult to find out really valuable information. ...

AI Technical Summary

Problems solved by technology

[0004] The current existing research on text sentiment analysis and sentiment classification only adds daily Internet terms, but ignores that specific words have different meanings in different contexts, and ignores specific contextual backgrounds

In addition, there are many short texts in the comment corpus of various social platforms, and the Chinese sentence structure is more casual, and the complex sentence structure also increases the difficulty of judging the emotional polarity of sentences.

Images