Research And System Implementation Of Sentiment Analysis Based On Grammar Rules

In contemporary society,Internet technology has seen leap development,in which case the network social platform,featured with convenience,originality and interactivity,gets popular among Internet users.As an increasing number of users express their opinions and emotions with the help of network social platforms,a large amount of text data with sentiment is brought,and how to obtain valuable information from a large amount of text data has become one of the hot spots of researchers.The task of text sentiment analysis is to divide the text into praise,derogatory,neutral and other sentiment categories according to the meaning and sentiment information expressed by the text,which is also termed as text orientation analysis because it is the division of the author's tendency and viewpoint.The core of traditional sentiment analysis methods lies in the construction of emotion dictionary and the selection of feature engineering.However,the construction of sentiment dictionary and feature engineering relies on a large number of manually annotated data.In recent years,with the application of neural network and other deep learning methods in the field of natural language processing,sentiment analysis method based on deep learning has been widely concerned in the field of sentiment analysis.However,the role of grammatical rules and syntactic structure is ignored by the traditional deep learning based sentiment analysis method,and the weak encoder is used to encode the text,which makes it difficult to improve the performance of the text sentiment analysis model.Aiming at the above problems,this paper proposes a method of integrating external knowledge and reconstructing emotion analysis framework,and applies the method to public opinion system to further verify the effectiveness of the method.The main work and innovation of this paper are as follows:(1)To solve the problem that sentiment analysis at text level and sentence level is not precise enough,this paper starts from fine-grained sentiment analysis and adopts aspect oriented emotional analysis method to identify the entities in sentences and analyze the emotional polarity corresponding to the entities to carry out more accurate text sentiment analysis.(2)To solve the problem that grammar rules are not fully used in current sentiment analysis methods,the grammatical rules are embedded into the model as external knowledge,and the emotional tendency of the text can be judged more accurately by combining the attention mechanism.(3)In order to solve the problem of weak encoder coding and ignoring syntactic structure in aspect-based sentiment analysis,an aspect oriented encoder is constructed to encode text,and graph convolution neural network is used to analyze sentiment of text.The experimental results show that this method can more accurately identify the aspects of emotion analysis.(4)The method of integrating external knowledge and reconstructing sentiment analysis framework is applied to the actual public opinion analysis system project,which further proves the practicability of the method.