The Application Of Data Mining In The Stock Analysis

Stock analysis is often dependent on a large number of stock analysis of historicaldata.How to analyze historical stock data anddig out valuable information is the majorityproblem of investors have been facing.Two stock analysis methods currently used isfundamental analysis and technical analysis.However,as the financial information in stockmarket accumulated more and more,the fundamental analysis and technical analysiscan noteffectively handle large-scale data.The traditional method is difficult to finduseful informationand valid rule from the large amount of data.Data mining technology provides an effectivewayfor the stock analysis.In this paper,using data mining technology to select stocks and select investmenttiming.This paper is divided into three main stages.First,analyze the key factots of stock byassociation rule analysis.Then use cluster analysis of the target stock based on the results ofassociation rules.The cluster analysisof stock resulting clustering stock for investors lookingfor stocks worth investing in a combination.Finally,the application of neural networkalgorithm to predict the index through experiments on different neural network algorithm tocompare and analyze the strengths and weaknesses of different algorithms andapplication.Experimental results show that the combination of RBF-BP neural networkalgorithm can better predict the trend of the socket, has some real meaning.The resultsprovide meaningful reference data for stock analysis and investment.Experimental results show that the application of data mining techniques to study thestock analysis to a certain extent to solve the shortcomings of traditional methods of stockanalysis. Data mining technology for stock is a valuable reference method.