一种不确定性条件下的自主式知识学习模型-软件学报.pdf

1000-9825/2003/14(06)1096 ©2003 Journal of Software 软 件 学 报 Vol.14, No.6 ∗ 一种不确定性条件下的自主式知识学习模型 + 王国胤 , 何 晓 (重庆邮电学院 计算机科学与技术研究所,重庆 400065) A Self-Learning Model under Uncertain Condition + WANG Guo-Yin , HE Xiao (Institute of Computer Science and Technology, Chongqing University of Posts and Telecommunications, Chongqing 400065, China) + Corresponding author: Phn: 86-23 Fax: 86-23 E-mail: wanggy@ Received 2002-06-03; Accepted 2002-11-06 Wang GY, He X. A self-learning model under uncertain condition. Journal of Software , 2003,14(6) : 1096~1102. /1000-9825/14/1096.htm Abstract: It is a very difficult problem in machine learning to learn uncertain knowledge automatically without prior domain knowledge. In this paper, a theory is developed to express, measure and process uncertain information and uncertain knowledge according to uncertainty measure of decision table and decision rule. Based on the Skowron’s default rule generation algorithm, a self-learning model and the method is developed to solve this problem. Simulation results illustrate the efficiency of this self-learning method. Key words: uncertainty ; rough set; self-learning; knowledge acquisition; machine learning 摘 要: 在没有领域先验知识条件下的不确定知识主动式学习是机器学习领域中的一个难题.通过研究决策 表和决策规则的不确定性,建立基于粗集表示、度量和处理不确定性信息和知识的理论,并且结合Skowron 的缺 省规则获取算法,提出一种不确定性条件下的数据自主式学习模型和方法,以解决这一问题.通过仿真实验,验证 了该自主式学习方法的有效性. 关键词: 不确定性;粗集; 自主式学习;知识获取;机器学习 中图法分类号: TP 18 文献标识码: A ∗ Supported by the National Natural Science Foundation of China under Grant No.698030 14 (国家自然科学基金); the National Climb Program of the Ministry of Science and Techn