多时间序列跨事务关联分析研究

论文的研究目的是为了对时间序列的发展趋势进行预测。采用的方法是对多时间序列进行跨事务关联规则分析,利用关联规则中前件和后件的时间差进行预测。提出了跨事务关联规则挖掘ITARM,该算法采用了基于压缩FP-树的、分而治之的挖掘方法。算法在产生了频繁1-项集之后,分别利用1-项集中的项作为约束条件,建立压缩FP-树,挖掘跨事务关联规则。文中给出了算法的主要设计思想和算法的伪代码,并对算法的性能进行了测试。测试结果表明,ITARM算法是一个时间和空间性能都较高的跨事务关联规则挖掘算法。

Research on Multiple Time Series Inter-transactional Association Analysis

The destination of this study is to predict the trend of time series.It adopts an approach with association rules analysis,and uses the time difference between the prerequisite and the consequent in an association rule to predict the trend.A new algorithm for inter-transactional association rules mining,ITARM,is presented.The algorithm uses a compact FP-tree based and divide-and-conquer approach.After the frequent 1-itemsets is produced,it separately uses them as constraint conditions to construct compact FP-tree and to mine inter-transactional association rules.It is introduced that the main idea and the pseudo-code of ITARM algorithm,and a performance test is done for the algorithm.The experimental results show that ITARM is an inter-transactional association rule mining algorithm with high temporal and spatial performance.