| 基于邻接网络的序列规则挖掘算法 |
| |
| 引用本文： | 陈富赞,寇纪淞,李敏强.基于邻接网络的序列规则挖掘算法[J].系统工程学报,2002,17(5):385-394. |
| |
| 作者姓名： | 陈富赞 寇纪淞 李敏强 |
| |
| 作者单位： | 天津大学管理学院,天津,300072 |
| |
| 基金项目： | 国家自然科学基金资助项目 (699740 2 6，70 1710 0 2 ) |
| |
| 摘 要： | 给出一个新的序列规则挖掘算法，该算法在挖掘规则以前将数据库预先存贮为序列邻接网络，在序列邻接网络中每个项目集顶点都有一个域来记录它的支持度，算法把频繁序列规则的发现问题转化为网络中的顶点搜索问题，大大提高了搜索过程的效率，为了有效地解决网络生成过程中的序列支持计算问题，采用了一种纵向的数据库表示格式。
|
| 关 键 词： | 邻接网络 序列规则 数据挖掘算法 关联规则 序列邻接网络 数据库 |
| 文章编号： | 1000-5781(2002)05-0385-10 |
| 修稿时间： | 2000年7月24日 |
Algorithm based on adjacent lattice for mining sequence rules |
| |
| CHEN Fu zan,KOU Ji song,LI Min qiang.Algorithm based on adjacent lattice for mining sequence rules[J].Journal of Systems Engineering,2002,17(5):385-394. |
| |
| Authors: | CHEN Fu zan KOU Ji song LI Min qiang |
| |
| Abstract: | A new algorithm for mining sequence rules is given in this paper. The algorithm converts the origin transaction database to sequences adjacent lattice in a preprocess, where each sequence vertex have a label to save its support. The algorithm changes the mining task in the database into a searching process in the lattice, which can speeds up greatly the mining. Furthermore, the algorithm uses a vertical id list database format to compute the support of each sequence, where each sequence is associated with a list of objects in which it occurs, along with the time stamps. |
| |
| Keywords: | data mining association rules sequence rules sequences adjacent lattice |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
