针对Apriori算法在频繁项集挖掘过程中的缺陷,提出了一种基于权重的改进Apriori算法.该改进算法通过一次扫描事务数据库构造出二元事务矩阵,再用各事务和各项的平均权重替代权重支持度,最终挖掘出事务库中的频繁项集.通过实例分析和性能测试,证明了改进的Apriori算法避免了重复扫描事务数据库,使得算法在性能上有了明显优化,并且挖掘出了Apriori挖掘不到的、隐藏的、有价值的规则.%An improved Apriori algorithm based on weight is proposed to deal with the defects of Apriori algorithm in frequent itemsets mining.In this improved algorithm,the binary transaction matrix is constructed through a scanning transaction database,and the weight support is then replaced with the average weight of each transaction and each item,and the frequent itemsets is finally digged out in the transaction library.Through the case analysis and performance test,it is proved that the improved Apriori algorithm can avoid the repeated scanning of the transaction database to make the algorithm optimized,and to excavate the hidden,hidden and valuable rules that Apriori cannot find before.
展开▼
