Sampling Based N-Hash Algorithm for Searching Frequent Itemset

Searching frequent itemsets is the critical problem in generating association rules in data mining, classic Hash-based technique, put forward by J. S. Park, for searching frequent itemsets has two shortcomings: one is that it is difficult to choose an appropriate hash function; the other is that it is liable to cause hash colliding. In order to solve the two problems, Chen Y.M. proposed N-Hash algorithm which needn't to choose hash function and avoided hash colliding. In this paper, the sampling technique is employed to improve the efficiency of N-Hash algorithm.