Language Expression and Economic Value

As the stock market becomes increasingly diverse and complex, an accurate feasible prediction of index has become urgent demand for investors. important driver changes in market, financial discourse can guide investors' emotions, thus affecting trading stocks development market. Therefore, from perspective emotion gradually a hotspot research. Through analysis existing literature, it is found that models used current relevant research are not ideal, accurate, there problems such as single method, few selection indicators, narrow area, etc. To solve above problems, this study proposes LSTM model integrating multiple feature emotional indexes, constructs TextCNN metaphorical power to quantitatively analyze expression semantic use news text, then integrates indexes into neural network predict Shanghai Stock Exchange Index.