摘要
本文给出了一种获取多类知识的决策树算法,该算法根据所给定的属性的优先级和取值类型进行分类型知识的获取。为了保证获得知识的有效性,根据科恩(Cohen)的归纳概率提出了一种证据支持程度来对所获得的知识进行评价,并相应地给出了一种知识求精的方法。
This paper gives a decision—tree algorithm acquiring many kinds ofknowledge.The algorithm depends on the given priority and type of attri-butes to acquire the classified type of knowledge.In order to ensure theeffectiveness of the acquired knowledge,a kind of evidence support degree toevaluate knowledge is proposed according to Cohen's inductive probability,and a kind of knowledge refinement method is given correspondingly.
出处
《国防科技大学学报》
EI
CAS
CSCD
北大核心
1990年第4期16-22,共7页
Journal of National University of Defense Technology
基金
国家自然科学基金资助
关键词
人工智能
知识获取
算法
决策树
artificial intelligence
knowledge acquisition
algorithm/decision—tree
knowledge evaluation
knowledge refinement
inductive probability
