Research On Investor Profile And Public Opinion Based On Guba Data

With the development of the Internet,Chinese investors are more and more enthusiastic about exchanging investment experience in forums,and the text information of stock reviews has also become a factor affecting the healthy development of the stock market.The rise and maturity of text mining technology in recent years has made it possible to realize investor-related research by mining stock review data.Based on the existing research results,taking the Guba website as the starting point,deep learning and machine learning methods are used to realize the classification of emotional sentiment in stock reviews and the identification of noise investors,and the relationship between public opinion and the stock market is studied.In the paper,crawling technology is used to obtain the required data in Guba.After data preprocessing,the Word2 Vec is used to implement the vectorization of words.In building Guba's sentiment dictionary,automatic tagging training corpus is realized by using the existing sentiment dictionary,and the neural network model is used to build the sentiment polarity classifier to expand the sentiment dictionary.It is found that the sentiment dictionary construction method used in this paper has a better effect on classifying sentiment tendencies in stock reviews than traditional methods.In identifying emotional investors,based on behavioral characteristics,a user clustering label is added containing the stock review information to establish a perceptual investor recognition model,which improves the effectiveness of the model.In order to study the relationship between public opinion and the stock market,the public opinion index is constructed based on the stock sentiment classification and sentiment investor identification research.On the one hand,the public opinion index establishes two VAR models with volatility and turnover rate respectively,and then Granger causality test are conducted.The results show that there is a positive correlation between public opinion and volatility,turnover rate,and two-way Granger causality.On the other hand,stock price prediction model is established,and results indicates that the model after adding public opinion index can better explain the changes in stock prices.All research results show that public opinion has an impact on the changes in the stock market,and there is a mutual stimulation between the public opinion and the volatility and turnover rate of the stock market.