Abstract: Apriori algorithm is a classical algorithm of association rules mining, in view of the deficiency of the Apriori algorithm some improvements are made. The new algorithm uses vertical data format, and improves the method of connecting the item to generate candidate. In order to study the stock plate cointegration phenomenon, the improved algorithm is applied to the stock plate index analysis. Experimental result shows that the improved algorithm can quickly find the corelation of the stock plate, it has a certain guiding role in the stock market analysis and investment decisions.

Key words: data mining, association rule, Apriori algorithm, stock analysis, stock plate indexes

摘要： Apriori算法是关联规则挖掘中的经典算法，针对Apriori算法的不足进行了一些改进。新算法使用垂直数据格式，并改进了产生候选项的连接方法。为了研究股票板块的联动关系，将改进算法应用于股票板块指数分析中。实验结果表明，改进算法能快速发现板块之间的联动关系，对股市分析和投资决策有一定的指导作用。

关键词: 数据挖掘, 关联规则, Apriori算法, 股票分析, 板块指数