| 一种改进的频繁项目集挖掘算法 |
| |
| 引用本文： | 张林,王宏海.一种改进的频繁项目集挖掘算法[J].阜阳师范学院学报(自然科学版),2013,30(1):54-57. |
| |
| 作者姓名： | 张林 王宏海 |
| |
| 作者单位： | 1. 安徽三联学院计算机科学与技术系,安徽合肥,230601
2. 安徽三联学院电子与通信技术系,安徽合肥,230601 |
| |
| 基金项目： | 安徽省高校省级自然科学研究项目 |
| |
| 摘 要： | 针对Apriori算法需要多次扫描数据库以及可能产生庞大候选集的瓶颈问题,提出了一种改进的频繁项目集挖掘算法,该算法仅通过一次数据库的扫描生成一个链表,以比特位的方式存储项目在事物数据库中出现的位置,并在不产生候选集的基础上通过逻辑运算与集合运算的直接生成频繁项目集。经过实例分析,结果表明该算法相对于Apriori算法,能够在保证准确率的基础上拥有更低廉的代价。
|
| 关 键 词： | 频繁项目集 数据挖掘 Apriori 算法 关联规则 |
An improved algorithm for frequent item sets mining |
| |
| ZHANG Lin
,
WANG Hong-hai.An improved algorithm for frequent item sets mining[J].Journal of Fuyang Teachers College:Natural Science,2013,30(1):54-57. |
| |
| Authors: | ZHANG Lin WANG Hong-hai |
| |
| Institution: | ( a. Department of Computer Science arm Technology, Department of Electronics and Communication Technology, Anhui Sanlian University, Hefei Anhui 230601, China) |
| |
| Abstract: | To solve the bottleneck problems that Apfiori algorithm needs to scan database time after time and may produce a large number of candidate sets, this paper proposes an improved algorithm for mining frequent item sets. This algorithm needed to scan database only once and generated a linked list whose node stored the position of item set in database by bit. Without producing candidate sets, the algorithm could directly generate frequent item sets by logic and set operation. The example analysis shows that compared with Apriori algorithm, this algorithm not only guarantees the accuracy but also costs less. |
| |
| Keywords: | frequent item sets data mining Apriori algorithm association roles |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
