基于GDT的不完整信息系统规则发现

提出了一个基于GDT的从不完整信息系统进行规则发现的方法。该方法利用GDT的思想，通过概括强度、规则置信度和规则强度等概念，充分考虑到数据不完整性和噪音引起的不确定性，在不改变原信息系统大小的前提下，直接从不完整信息系统中得到简洁实用的规则。最后，通过一个例子阐述了该方法的实施过程，并将该方法与提及的其它几种不从不完整信息系统中发现规则的方法进行了分析比较。分析表明该方法是一种新的有效的从不完整信息系统抽取规则的途径。

RULES DISCOVERY FROM INCOMPLETE INFORMATION SYSTEMS BY GDT APPROACH

In this paper, a GDT based approach for rules discovery from incomplete information systems is proposed. This approach introduces the GDT technique to extract rules with strength so that the rule could effectively perform in real application. The approach does not require changing the size of original incomplete information systems and is suitable for any information systems. Finally, several existing methods of computing decision rules from incomplete information systems are analyzed and compared. The results show that this approach is a new effective approach for rule discovery from incomplete information system.