中文说明： 关联规则挖掘是从事务集合中挖掘出这样的关联规则：它的支持度和置信度大于最低阈值（minsup,minconf），这个阈值是由用户指定的。根据支持度=(X,Y).count/T.count，置信度=(X,Y).count/X.count ，要想找出满足条件的关联规则，首先必须找出这样的集合F=X U Y ，它满足F.count/T.count ≥ minsup，其中F.count是T中包含F的事务的个数，然后再从F中找出这样的蕴含式X—>Y，它满足(X,Y).count/X.count ≥ minconf，并且X=F-Y。我们称像F这样的集合称为频繁项目集，假如F中的元素个数为k，我们称这样的频繁项目集为k-频繁项目集，它是项目集合I的子集。所以关联规则挖掘可以大致分为两步： 1）从事务集合中找出频繁项目集； 2）从频繁项目集合中生成满足最低置信度的关联规则。
English Description：
Association rule mining is to mine such association rules from the transaction set: its support and confidence are greater than the minimum threshold (minsup, minconf), which is specified by the user. According to the support = (x, y). Count / t.count, confidence = (x, y). Count / x.count & nbsp;, in order to find the association rules satisfying the conditions, we must first find such a set F = x U & nbsp; y, which satisfies f.count / t.count & nbsp; ≥ minsup, where f.count is the number of transactions in t containing F, and then find such an implication X - & gt; y from F, which satisfies (x, y). Count / x.count & nbsp; ≥ minsup Minconf, and x = f-y. We call a set like f a frequent itemset. If the number of elements in F is k, we call such a frequent itemset k-frequent itemset, which is a subset of itemset I. So the mining of association rules can be divided into two steps