摘要
关联规则是数据挖掘的重要内容之一。为了高效、快速地从事务数据库中挖掘出频繁项目集,针对数据挖掘的经典关联规则Apriori算法的瓶颈问题提出了改进的方法。通过对基于数组的Apriori算法的改进,只扫描一次数据库,在生成候选频繁项目集前进行判断,减少非频繁的候选的项目集的生成,并通过减少数组数据的扫描和不断压缩数组,提高了算法的运行效率,节约了开销。
The associated rules is one important part of data mining. In order to efficiently and rapidly find frequent set of items from the database. Against to the associated rules apriori Algorithm bottlenecks questions improved. It is based on the number of improvements Apriori algorithm, only one database of scanned, judgeing before generating candidate frequent set of items, reduced to set of items of data to reduce the number of scans and improve the efficiency of the algorithm run- ning. It is a saving of cost.
出处
《计算机与数字工程》
2011年第8期1-3,24,共4页
Computer & Digital Engineering
基金
广西研究生教育创新计划自助项目资助