摘要
Apriori算法是发现频繁项目集的经典算法,但是该算法需反复扫描数据库,因此效率较低。针对Apriori算法,GBARM (Group Based Association Rules Mining)算法对事务集进行压缩,并且在统计Ck中各项集的支持频度时,逐步减小Ck的规模,从而改善算法的性能。
Apriori is a classical algorithm of generationg frequent itemsets. But the algorithm need repeatedly scan the transaction database, and the performance will be dramatically affected. The GBRAM （Group_Based AssoCiation Rules Mining） algorithm is albe to gradually reduce the scale of the transaction used to scanning and gradually reduce the scale of the Ck at the same time. As a result, the performance of the algorithm is significantly improved.
