基于SQL Server 2000下数据挖掘算法的研究  

Research of data mining algorithm based on SQL Server 2000

作　　者：王洪立[1] 吴继娟[1] 

机构地区：[1]哈尔滨工程大学计算机科学与技术学院,黑龙江哈尔滨150001

出　　处：《计算机工程与设计》2008年第3期759-761,共3页Computer Engineering and Design

摘　　要：微软的SQL Server2000是当今最流行的数据库管理软件之一,研究了在SQL Server 2000上数据挖掘实现方面的决策树算法。决策树算法通过构造精度高、小规模的决策树采掘训练集中的分类知识。SQL Server 2000/Analysis Service两层结构决策树,采用了以类记数表及深度优先策略生成,在建树算法和数据库间设立数据挖掘中间件。并讨论了通过使用像SQL Server 2000 Analysis Service这样的典型工具来如何实现数据挖掘模型的创建,且为商业组织的决定挖掘出必要的数据。Microsoft SQL Server 2000 is being one of the most popular and big software for managing databases. The decision tree algorithm is investigated that the main methodology in the implementation of data mining on SQL Server 2000. Decision tree algorithm is that the category knowledge of the training set is mined through built high precision and small-scale decision tree. Two layers decision tree algorithm of SQL Server 2000/Analysis Service is created by class count table and depth In＇st strategy. A data mining middleware is set up between the algorithm built tree and database. Therefore, this paper discourses that through the usage of typical tool like SQL Server 2000 Analysis Services. How data mining model can successfully be created and leveraged as a next step towards uncovering and discovering essential data for business decision making within a business organization.

关 键 词：数据挖掘 微软SQL SERVER2000 分析服务 决策树算法 数据挖掘模型 

分 类 号：TP311.131[自动化与计算机技术—计算机软件与理论]