Association Rule Algorithm In Stock Market's Application

In the last 1980's, data mining—a new research field, appears gradually. The study purpose of data mining is to find the regular information, which is hided in large data set and people are interested in. In this paper, we only focus on an important field of data mining—association rules analyse. In 1993, Agrawal R proposed the association rules problem. Henceforth, association rules mining obtained extensive attention of numerous experts and scholars, and it became an important branch of data mining.Firstly, this paper introduces association rule and the Apriori algorithm and analyzes existent flaw of the Apriori algorithm. Secondly, this paper analyzes existent question of The Apriori algorithm in the stock market, points out as a result of stock market's particularity, the Apriori algorithm cannot effective carry on the stock analysis. This paper applies ABC analysis to the analysis of stock analysis, sorts out the different types of stocks and introduces two kind of alterant Apriori algorithms: Based on new transaction behavior sensitive Apriori optimization algorithm and based on weight parameter Apriori optimization algorithm. This paper proposes the vertical weighted computation support and a horizontal weighting computation support and weight computational method based on the Apriori algorithm and Prove the vertical weighted support applicable to the traditional Apriori algorithm, we use to stock analysis, According to the nature of Apriori algorithm define the weight between itemset and the definition of support. In the definition of weight of individual item, we can combine to ABC analysis to adjust the weight of individual item, consider frequency and importance of the two. In the fourth chapter, we carry through stock analysis, pretreatment of the data, the processing of stock code, the transaction time and data conversion, we analysis stock, According Apriori algorithm and improved algorithms.