摘要
文章介绍了一种增量式关联规则更新方法,其核心思想是,将长事务以时间分割,分成一个连续的情节集合,当前情节期间获得的信息,依赖于当前的事务子集以及前面情节期间已经发现的信息。仅使用更新的事务和前面阶段的挖掘结果,增量式地产生频集。用Apriori类算法作为局部过程来产生频集,给出了具体的动态挖掘算法。
This paper proposes an incremental updating approach based on time divided transactions for mining association rules,which discovers current data mining rules by using updates that have occurred during the current episode along with the data mining rules that have been discovered in the previous episode.An Apriori-like approach as a local procedure is used to generate large itemsets,and the detail algorithm is given.
出处
《计算机工程与应用》
CSCD
北大核心
2004年第23期176-179,共4页
Computer Engineering and Applications