一种结合完全连接的改进Apriori算法

Improved Apriori algorithm based on the absolutely join

基于Apriori算法原理,提出一种有效的完全连接条件,在频繁2k项集的集合L2k进行自身Apriori连接得频繁(2k+1)项集的同时,自身完全连接产生未剪枝的候选4k项集;对频繁(2k+1)项集的集合L2k+1,直接对其项集进行完全连接产生未剪枝的候选(4k+2)项集。改进的算法减少了连接的比较次数、迭代运算次数。实验表明该算法在保证无遗漏的情况下有效地提高了Apriori算法的挖掘速度。

Based on the principle of Apriori, a excellent premiss of absolutely join was presented, the candidate 4k-itemsets were built directly with absolutely join while created the candidate （2k ＋ 1） -itemsets from L2k （ the muster of frequent 2k-itemsets） ; and only used the absolutely join for L2k＋1（ the muster of frequent （2k ＋ 1）-itemsets） to create the candidate （4k ＋ 2）-itemsets. This algorithm decreases the times of iteration and the compare. The experiment results show that no frequent itemsets is missed and the speed of the mining is effectively improved in this algorithm.