非精确概念的机器学习与应用  

Machine learning and application of inaccurate concepts

作　　者：韩习武[1] 赵铁军[1] 

机构地区：[1]哈尔滨工业大学计算机科学与技术学院,哈尔滨150001

出　　处：《哈尔滨工业大学学报》2006年第10期1736-1739,共4页Journal of Harbin Institute of Technology

基　　金：国家自然科学基金资助项目(60373101)

摘　　要：提出并描述非精确概念的机器学习问题,定义非精确概念学习的一般任务模式,给出符合人类学习习惯的解决方案.结合具体例子说明非精确概念的学习与应用的过程,总结非精确概念学习同时具有归纳学习和统计学习的、优势及很强的鲁棒性,其缺点是计算量大,需要维扩一个很大的目标函数假设集合及没能充分考虑领域知识.并指出本文学习方法的可能应用领域.In practical tasks concept learning often involves a process from inaccurateness to accurateness. This paper describes the machine learning problem of inaccurate concepts, defines a general model of the inaccurate concept learning task, and proposes a possible solution, which suits relevant human learning habits. Then the process of inaccurate concept learning and application is illustrated with a concrete example. At last, features, advantages and limits are discussed, with possible application fields indicated.

关 键 词：非精确概念 机器学习 任务模式 

分 类 号：TP181[自动化与计算机技术—控制理论与控制工程]