一种多关系频繁模式挖掘算法

Multi-relational frequent pattern mining algorithm

传统数据挖掘算法在处理多表时,需要物理连接,存在效率不高的问题。为了解决这一问题,提出了一种多关系频繁模式挖掘算法。该算法利用元组ID传播的思想,使多表间无须物理连接,就可以直接挖掘频繁模式。实验表明,此算法具有较高的效率。

While dealing with multi-relation, traditional data mining algorithms used the way of physical join. In order to solve this problem, this paper proposed a multi-relational frequent pattern mining algorithm. By taking advantage of tuple ID propagation approach, this algorithm could directly mine frequent pattern in multi-relation without physical join. Experiment demonstrates that, this algorithm has high efficiency.