Affiliations: Hebei University of Engineering, Handan, Hebei, China
Correspondence:
[*]
Corresponding author. Linfeng Guo, Hebei University of Engineering, Handan, Hebei, China. E-mail: MeganSSyN@yahoo.com.
Abstract: Stock market trading is relatively difficult due to the relatively complicated investment in financial markets. In order to solve this problem, an industry configuration model based on deep learning network sentiment mining and fundamental research is proposed innovatively in this paper. Firstly, by crawling the network information, the network public opinion data is used to conduct industry classification and sentiment analysis on the data, thereby obtaining the industry sentiment index and the industry income forecast. Then combined with the fundamental data and market data and the technical indicators generated by it, multi-factor analysis is carried out, and the income forecast is obtained. Through three sets of comparative experiments, the results show that the investment returns obtained are the best with the industry configuration model based on deep learning-based network sentiment mining and fundamental research.
Keywords: Recurrent neural network, internet emotion mining, word vector analysis, industry configuration, fundamental analysis
