一种高效的关联规则连续增量更新改进算法

An Efficient Incremental Updating Algorithm for Association Rules

针对FP-growth算法存在的不能进行增量更新,以及已有基于FP-growth的增量更新算法效率不高、不支持连续更新等问题,在FP-tree基础上,提出了增量更新改进算法FPIUA2,其适用于数据集连续增加的情形,适用于稀疏型数据集和稠密型数据集、支持连续执行.实验表明:该算法的效率远高于FPgrowth和已有的增量更新算法,其执行效率较FP-Growth、FPUA和FIUA2算法提高了1个数量级,并且具有很好的可扩展性.

In order to overcome the shortage of FP - growth algorithm and the improved algorithms basedon the FP - growth strategy, three updating algorithms FPIUA2 which based on the EFP - tree are proposed.It＇ s very fast in the case of the data set continually being added . Both theoretical and practical analysisshow that our new algorithms are better than existed algorithms and are suitable for both sparse and densedata set. Thus it is about an order of magnitude faster than the FP - Growth algorithm, FPUA algorithm andFIUA2 algorithm.