| 基于前缀的Apriori算法 |
| |
| 引用本文： | 粟莉萍,杨文伟.基于前缀的Apriori算法[J].微型机与应用,2011,30(4):75-78. |
| |
| 作者姓名： | 粟莉萍 杨文伟 |
| |
| 作者单位： | 广东工业大学计算机学院,广州广州,510006 |
| |
| 摘 要： | 通过对Apriori算法的研究和分析,结合算法存在的缺陷,利用"桶"技术及压缩组合项集技术,对频繁项集提出了前缀概念,并提出了基于前缀的频繁项集挖掘算法。该算法将具有同一前缀的频繁项集的子集合作为一个节点,由频繁k-项集的子集合直接产生候选(k+1)-项集,从而省略了连接步中判断I1、I2是否能连接。同时,该算法使得整个程序中节点数目减少,这样不仅减少了内存消耗,而且提高了查找Ck和Lk的速度,尤其便于大型数据库的分布式处理。经实验证实,改进后的算法是可行的。
|
| 关 键 词： | Apriori算法 关联规则 频繁项集 “桶”技术 压缩组合技术 |
Apriori algorithm based on prefix |
| |
| Su Liping,Yang Wenwei.Apriori algorithm based on prefix[J].Microcomputer & its Applications,2011,30(4):75-78. |
| |
| Authors: | Su Liping Yang Wenwei |
| |
| Affiliation: | Su Liping,Yang Wenwei(College of Computer,Guangdong University of Technology,Guangzhou 510006,China) |
| |
| Abstract: | In view of Apriori algorithm research and analysis,combined with the existence of the algorithm,according to Bucket technology and compression technology,the paper puts forward the prefix concept to frequent itemsets and frequent itemsets mining algorithm based on prefix.The new algorithm lets the subset of frequent itemsets having a same prefix as a node,and has a decrease in number of nodes,thus reduces the storage space and improves the speed of finding Ck or Lk.Through the node generates candidate K-fre... |
| |
| Keywords: | Apriori algorithm association rules frequent itemsets Bucket technology compression technology |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
