计算机工程与应用 ›› 2011, Vol. 47 ›› Issue (7): 153-155.

• 数据库、信号与信息处理 • 上一篇    下一篇

基于邻接矩阵的FP-tree构造算法

刘应东1,冷明伟2,陈晓云3   

  1. 1.兰州交通大学 交通运输学院,兰州 730070
    2.上饶师范学院 数学与计算机系,江西 上饶 334000
    3.兰州大学 信息科学与工程学院,兰州 730000
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2011-03-01 发布日期:2011-03-01

Construction algorithm of FP-tree based on adjacency matrix

LIU Yingdong1,LENG Mingwei2,CHEN Xiaoyun3   

  1. 1.School of Traffic and Transportation,Lanzhou Jiaotong University,Lanzhou 730070,China
    2.Department of Mathematics and Computer,Shangrao Normal University,Shangrao,Jiangxi 334000,China
    3.School of Information Science and Engineering,Lanzhou University,Lanzhou 730000,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2011-03-01 Published:2011-03-01

摘要: 提出了一种基于邻接矩阵的FP-tree构造方法。首先通过扫描数据库建立2-项集支持数的邻接矩阵,通过邻接矩阵对项进行过滤和新方式排序,然后再利用邻接矩阵构造FP-tree,使得FP-tree的分支、节点数和深度大幅度地减少,从而使存储空间减少、遍历时间缩短。最后使用标准数据集进行验证测试并和其他算法的比较,实验结果表明,该算法在保证结果的同时有效地提高频繁项集挖掘的效率。

关键词: 数据挖掘, 频繁项集, FP-tree算法, 邻接矩阵

Abstract: A construction algorithm of FP-tree based on adjacency matrix is proposed.An adjacency matrix about support count of 2-frequent item sets is constructed by scanning database.Using the adjacency matrix,FP-tree is established after item sets are filtered and restructured.For the numbers of branches,nodes and depths are reduced greatly,the storage space is far less and ergodic time is shorter much.The construction algorithm is tested and verified using standard datasets.The result shows the new construction strategy can improve efficiency of frequent item mining and ensure validity of the results compared with others algorithms.

Key words: data mining, frequent itemsets, FP-tree algorithm, adjacency matrix