频繁项集的简洁表示方法研究

提出了通过逻辑蕴涵的演绎规则来构建频繁项集的简洁表示,而不是挖掘所有的频繁项,来降低频繁项集中的冗余性,并利用演绎规则来构建所有频繁项集的简洁表示,从而减少存取数据库和对项集进行计数的次数,最终提供一个更加紧凑的频繁项集输出.最后,通过试验说明了该方法的有效性.

Research on Methods of Concise Representation for Frequent Itemsets

In this paper, a new method for constructing concise representation of frequent itemsets based on deduction rule is proposed. Lower and upper bounds on the frequency of a candidate are calculated with deduction rules, based on the frequencies of all its subsets, the rules derive best possible bounds on the frequency of an itemset. In this way, the number of candidate itemsets to count can be avoided with these bounds. Therefore, it is possible to generate a summary of the frequent itemsets that only contains the non-derivable sets, then the collection of none-derivable sets forms a concise representation of the frequent itemset. The experimental results show the effectiveness and efficiency of this method on a real-life dataset.