The Application Of Data Mining Technology In Stock Analysis And Forecast

In recent years,data mining technology has been widely applied in various fields.It has an overwhelming advantage in handling massive data and Knowledge Discovery. There are lots of hidden disciplines exist in tremendous amount of data about stock transactions which can't be easily discovered by ourselves for human's capability in Data Processing.However,data mining technology can make up for this deficiency.This paper attempts to use data mining technology in stock analysis and forecast.First over all,this paper introduces the background knowledge and methods about stock analysis and forecast.Then do a thorough and meticulous research into the theory of data mining.At last discuss the following three aspects:First of all,focus on the research into financial indexes of stock.Then select some representative ones which can be analyzed in use of algorithm ID3 of Decision Tree Classification. Investors can use the test results to analyze the operating conditions and profitability of listed companies.Then Make use of Association Rules to discover the linkages between stocks. Mining association rules in transactional database can find rules among items.However,the data of stock is not boolean but series.So a new method E-QA is used to discover the association rules among stocks.At last,Use algorithm of time series similarity search to find the similary history stock data with the given stock's direction.The stock's directions always has many rules and paterns, and these paterns often present itself. If we can find the similary history data with the current stock direction, we can select invest's strategy according to it.