Abstract:
Tax date has a large mount and FP_TREE algorithm needs huge memory for data mining. To solve these problems, an association rule algorithm based on binary tree frequency pastern tree is announced. The algorithm stores data by binary tree to decrease the time that accessing database. It reduces the memory needed by pre-build the binary pastern of the binary pastern. It is applied to the analysis of the data of taxation law enforcement, and the potential rules of faults accrued during taxation law enforcement are found out. It raises the scientificalness and actual effect of management for taxation law enforcement.
Key words:
data mining,
association rule,
taxation law enforcement
摘要: 针对税收执法数据量大和频繁模式树FP_TREE算法在挖掘海量数据时需要占用大量内存的缺点,提出一种基于二叉频繁模式树FP_Btree的关联规则算法。算法用二叉树存储数据,减少对数据库的访问次数。采用先求出先建立的二叉频繁模式树的频繁模式,减少算法的内存占用量。该算法已应用于某市税收执法数据分析中,能找出执法过错行为的潜在规律,提高税收执法管理的科学性、实效性。
关键词:
数据挖掘,
关联规则,
税收执法
CLC Number:
YAO Liang; XU Shao-bin; HU Xue-gang;. Application of Association Rule Mining in Management of Taxation Law Enforcement[J]. Computer Engineering, 2008, 34(24): 266-267.
姚 亮;徐邵兵;胡学钢;. 关联规则挖掘在税收执法管理中的应用[J]. 计算机工程, 2008, 34(24): 266-267.