一种新型的改进Apriori算法研究  

An improved apriori algorithm

作　　者：白东玲[1] 郭绍永[2] 王晓[1] 贺新禹[1] 

机构地区：[1]新乡医学院计算机中心,河南新乡453003 [2]新乡医学院现代教育技术中心,河南新乡453003

出　　处：《信息技术》2013年第7期50-53,共4页Information Technology

基　　金：2012年河南省社科联;经团联调研课题(SKL-2012-719)

摘　　要：Apriori算法是数据挖掘关联规则研究中的经典算法,由于它需要多次扫描数据库,造成系统运行效率比较低。所以在对Apriori算法进行了分析之后提出了改进的Apriori算法。改进的主要思想是基于将事务数据库转化成相应的0-1矩阵,通过对矩阵中每个向量与其后的向量做内积运算来计算支持度,并将计算得到的支持度与给定的最小支持度作比较,删除小于最小支持度的行与列,缩小矩阵的规模,提高了运行速度效率。改进的算法只需要对数据库扫描一次,运行效率比较高。实验结果表明,该优化方案是有效可行的。Apriori algorithm is a classical algorithm in data mining association rules. However the algorithm scans the database so many times causes that it runs too slowly. A new Apriori algorithm is proposed after Apriori is analyzed. The main idea of the new Apriori algorithm is that transaction database is converted to corresponding 0-1 matrix, whose each vector and subsequent vector does inner product operation to get support. Comparing with the given minsupport, it deletes the rows and columns if they are less than the minsupport to reducing the size of the rating matrix. It makes the computing more efficient paper It runs slowly as it scans the database frequently. The improved algorithm only needs to scan the database once so that it runs quickly. The experiment results prove that it is efficient and feasible for the improved algorithm.

关 键 词：矩阵 关联规则 改进Apriori算法 数据挖掘 

分 类 号：TP311.13[自动化与计算机技术—计算机软件与理论]