Analysis Of Stock Selection Based On Data Mining Technology

China's stock market has experienced nearly three decades of development,with the stock market system gradually improved,more and more investors have involved in stock investment.However,because of the high risk of the stock market,it is important to find a simple and easy way for investors.The stock price is affected by many factors,it is superficial to rely on artificial or simple statistics analysis to make investment decisions,and data mining technology is a good way to solve this problem.Data mining technologies extract effective information that is not found but used for decision support by the means of dealing with data with large volume and high complexity.The application of data mining technology to stock investment is helpful to reduce the investment risk of investors,improve the investment decision and evaluate the investment value of securities correctly.This paper used three kinds of data mining technology to find the stocks which have the investment value.This paper used the association rules,decision tree model and neural network model to create a model,while used the financial indicators,internal situation indicators and the external industry indicators as input variables,the stock rate of return over the broader market yield as the dependent variable,It has found that the two models of decision tree and neural network have prediction effect in improving stock picking ability.There is a relationship between financial indicators,internal situation indicators,external industry indicators and stock changes.Neural network model has no obvious advantages than decision tree model in this paper.