一种基于压缩矩阵的Apriori算法改进研究.pdfVIP

Vol. 40 No . 12 第 40 卷第12 期 计算机科学 2013 年 12 月 Computer Science Dec 2013 一种基于压缩矩阵的 Apriori 算法改进研究 罗丹李陶深 (广西大学计算机与电子信息学院 南宁 530004) 摘 要 针对已有基于短碎的 Apriori:算法存在的问题，提出了一种改进的基于压缩短碎的 Apriori 算法。算法进行 了以下方面的改进: 增加了两个数组，分别用于记录矩阵行与列中 1 的个数，使得算法在压缩短阵时减少了扫描矩碎 的次数;在压缩短阵中，通过增加州除不能连接的项集和非频繁的项集的操作，使得短阵压缩得史小，提高了空间效 率;改变了删除事务列的条件和算法结束的条件，以减少挖掘给采的误差和算法循环的次数。算法性能分析和实验分 析证明，改进后的算法能有效地挖掘频繁项集，并且比现有的算法具有更高的计算效率。 关键词 数据挖掘，频繁项集.Apriori :算法，压缩短阵 中图法分类号 T P31 1 文献标识码 A R臼铺rch 00 Improved Apriori Algorithm Based on C佣lpr咽sed Matrix LUO Dan LI Tao-shen (Sch∞1 of Computer t Eleçtronics and Informationt Guangxi University. Nanning 530004t China) Abstr在ct Aiming at the deficiency of the existing Apriori algorithm , an improved Apriori algorithm based on ∞m­ pressed matrix called NCM_Apriori_l was proposed. The improvements of this algorithm are as follows: (1) adding two arrays to record the ∞unts of 1 in the row and column ,so that the number of scanning the matrix can be reduced during compress吨，(2)deleting the unnecessary itemsets which can t be connected as well as the infrequent ones in compress帽 ing matrix to minify the scaJ e of matrix and improve space utilization , (3)changing the ∞ndition of deleting the unneces- sary transactions tO reduce the e汀ors of the mine result ,and changing the sωpping condition to make the numþer of cy- cle decreased. Algorithm performance analysis and experiments results prove that the improved algorithrηcan mine fre- Quent itemsets effωlively and has better efficiency of computing than existing Apriori algorithms based on compressed