摘要
在数据挖掘中使用本体和上下文知识能够将普遍的知识和特定的知识引入数据挖掘的决策因素中,是增进数据挖掘准确性的有效手段,同时也是数据挖掘领域研究的热点和难点之一.针对该问题,首先探讨了本体与上下文知识的集成化表示方法,包括上下文知识分类方法、如何在本体描述方法上扩展上下文知识及上下文知识转化方法.其次,以层次化结构的本体与上下文知识为例,构建了一个依据于本体和上下文知识集成的归纳学习算法并验证了该算法的有效性和准确性.
Using ontology and context knowledge in data mining is one of the effective waies to improve data mining accurateness, which can add general knowledge and certain knowledge in decision factors. How to apply ontology and context knowledge in data mining is discussed in this paper. Firstly, the integration model of ontology and context knowledge is presented, which includes context information categories, context information extended on ontology models and context transformation method. Based on those, using the hierarchy structure of the ontology and context knowledge integration model as an example, the induced learning algorithm is presented in terms of the integration ontology and context knowledge. The experiment of the induced learning is presented and its result is more effective and accurate.
出处
《软件学报》
EI
CSCD
北大核心
2007年第10期2507-2515,共9页
Journal of Software
基金
Supported by the National High-Tech Research and Development Plan of China under Grant No.2007AA04Z148 (国家高技术研究发 展计划(863))
the National Natural Science Foundation of China under Grant No.60573126 (国家自然科学基金)
the National Basic Research Program of China under Grant No.2002CB312005 (国家重点基础研究发展计划(973))
关键词
数据挖掘
本体
上下文知识
归纳学习算法
data mining
ontology
context knowledge
induced learning algorithm