MiningAlgorithmsofN-MostFrequentItemsets

说明：  频繁项集挖掘算法的计算复杂性和生成的频繁项集数量随着事务集项数的增加呈指数增长，最小支持度阈值成为控制这种增长的关键．然而，实际应用中仅使用支持度阈值难以有效控制频繁项集的规模．为此定义N个 最频繁项集挖掘问题，并提出基于支持度阈值动态调整策略的宽度优先搜索算法Apriori和深度优先搜索算法IntvMatrix挖掘N个最频繁项集．实验表明，本文的2种方法的效率比朴素方法高2倍以上，特别当N值较低时，本 文方法的效率优势更为明显．
(Frequent itemsets mining algorithm for calculating the complexity and frequent itemsets generated by the number of sets with the affairs of an exponential increase in the number of growth, minimum support threshold to become the key to control this growth. However, the practical application using only support threshold difficult to effectively control the scale of frequent itemsets. For this reason the definition of N most frequent itemsets mining problem and based on support for dynamic adjustment of the threshold strategy-first search algorithm Apriori width and depth-first search algorithm IntvMatrix Mining N most frequent itemsets. Experiments show that the two kinds of methods than the simple method of high-efficiency 2 times more, particularly when the N value is low, the efficiency advantages of this method is more obvious.)