摘要
数据挖掘已成为信息时代获取知识的重要手段。通过数据挖掘获取的知识具有大量性、新颖性、粗糙性、时效性等特点,传统的知识评价手段难以对这类知识进行快速、有效地评价,因而产生知识冲突、知识过载等问题。利用可拓学理论、人工智能和复杂性科学理论,研究知识本身具有记忆、识别、聚集等智能特点的机理,提出知识在应用过程中进行自我评价的指标及方法,为数据挖掘获取的知识的智能化评价提出一条新路径,对丰富知识审计的研究内容、促进数据挖掘的实践应用具有理论指导意义。
Data mining has become an important mean for acquiring knowledge.Since the knowledge obtained by data mining methods is huge,novel,rough,and fast-changing,it goes beyond the capability of traditional knowledge auditing methods.Thus,the knowledge conflicting and knowledge overloading have occurred.The knowledge mechanism with a series of intelligent features,such as memory,recognition and clustering is explored by the extension theory(extenics) and complexity theory.Then a new method and its KPIs which knowledge can self-evaluate during the application process are presented.A new path for the intelligent evaluation of knowledge is provided;it enriches the content of knowledge audit and has the theoretical significance for promoting the practice of data mining.
出处
《科研管理》
CSSCI
北大核心
2010年第S1期32-38,共7页
Science Research Management
基金
国家自然科学基金:数据挖掘获取的知识的智能化管理研究(项目号70871111
2009.1-2011.12)
国家自然科学基金委创新研究群体科学基金"数据挖掘与智能知识管理:理论及应用研究"(项目号70621001
2007-2012)
宁波市软科学基金:金融危机背景下宁波企业技术创新战略与对策研究(项目号2009A10033
2009.3-2010.6)
甘肃省科技支撑计划项目:面向中小企业的自主创新平台系统(项目号090GKCA030
2009.4-2011.4)
关键词
知识评价
数据挖掘
智能知识管理
可拓学
knowledge audit
data mining
intelligent knowledge management
extenics
