基于先验知识的缺损数据研究  

On Imperfect Data of IMS Based on Experience Knowledge

作　　者：娄海彦[1] 楼巍[1] 严利民[1] 

机构地区：[1]上海大学自动化系,上海200072

出　　处：《控制工程》2004年第6期539-541,共3页Control Engineering of China

摘　　要：缺损数据是企业在实现信息化管理过程中需要重视和处理的重点问题之一。以粮食企业信息化管理系统为实际操作的样本,在其关系型数据库中,属性之间的相互关联关系(称之为先验知识或领域知识)为修正和填充缺损数据提供了大量有益的帮助。采用了基于先验知识的缺损数据处理方法,重点针对系统中的离散属性的缺损数据进行了研究,对该系统中的缺损数据的处理有着很强的实用性,并为其他类似问题的解决提供了可借鉴的方案。The processing of imperfect data is one of the important problems that need to be specially considered in the realization of information management system.The foodstuff enterprise information management system is taken as a practical model.For the RDS,the associated relationship among the properties (called experience knowledge or field knowledge) helps to modify and fill the missing values of the system.A method based on experience knowledge with good practicability is used for this system.It provides a kind of way for reference to solve those similar problems.

关 键 词：信息管理系统 缺损数据 先验知识 信息化管理 

分 类 号：TP274[自动化与计算机技术—检测技术与自动化装置]