摘要
提出了一种标准粗糙集约简时连续属性离散化的新方法。采用标准粗糙集进行属性约简时,要求属性为离散的,而大多数情况下属性是连续的,因此需要进行离散化处理。首先介绍了原有的信息熵算法并指出其局限性;其次,对多类别信息熵进行扩充,将距离因素引入到该信息熵的计算中;最后给出了扩展信息熵计算的两个基本准则,利用证据理论完成信度的上聚焦。仿真显示了该方法的有效性。
A new discretized approach for continuous attributes was presented. As we know, the basic rough set theory can not deal with continuous attributes, so the continuous attributes need to be discretization. Firstly, the disadvantages of original information entropy were analyzed. Secondly, the distance factor was introduced to expand the information entropy. At last two basic rules were presented to calculate expanded information entropy and DS evidential theory was used to process the mass up﹣focus. Simulation results show the availability of this new information fusion algorithm.
出处
《红外与激光工程》
EI
CSCD
北大核心
2014年第11期3802-3806,共5页
Infrared and Laser Engineering
基金
河南省科技攻关重点计划项目(122102210563
132102210215)
关键词
粗糙集理论
离散化
属性约简
连续属性
证据理论
rough set theory
discretization
attribute reduct
continuous attributes
evidential theory
