案例知识库维护技术的研究进展  

Some advances in maintenance research of case knowledge base

作　　者：李建洋[1] 倪志伟[1] 郑金彬[2] 谢秀珍[2] 

机构地区：[1]合肥工业大学计算机网络所,安徽合肥230009 [2]龙岩学院计算机科学系,福建龙岩364000

出　　处：《武汉工程大学学报》2010年第3期96-99,共4页Journal of Wuhan Institute of Technology

基　　金：863专项课题(2007AA04Z116);国家自然基金(70871033);福建省科技专项课题(2008F5043);安徽省高校自然基金(KJ2008B107);福建省教育科研基金(JA08229)

摘　　要：案例库是CBR系统的主要知识库,但是难以维护,一个主要因素是案例库大,并且是非结构化或半结构化的,用自然语言来表达的.文章针对案例知识库维护中出现的相关复杂问题,分析了各种维护策略的可行性,提出了在不同环境下实施维护的准则,指出了选用合适的方法来实现案例库维护.With the dramatic proliferation of Case-Based Reasoning （CBR） systems in commercial applications, CBR has grown from a quite new area to a subject of major influence within Artificial Intelligence. Case bases are main knowledge sources in CBR systems, but they are difficult to maintain. One of the contributing factors is that these case bases are often large and yet unstructured or semi- structured, they are represented in natural language text. Tackling the following problems related to the complexity of case base maintenance （CBM）, the paper presents many kinds of strategies to achieve CBM and puts forward some criteria for this course, it suggests choosing the most suitable method for CBM based on its feasibility, which are under description in the face of different conditions.

关 键 词：案例库维护 不确定删除法 选择删除法 粒度计算 

分 类 号：TP39[自动化与计算机技术—计算机应用技术] TP181[自动化与计算机技术—计算机科学与技术]