检索规则说明:AND代表“并且”;OR代表“或者”;NOT代表“不包含”;(注意必须大写,运算符两边需空一格)
检 索 范 例 :范例一: (K=图书馆学 OR K=情报学) AND A=范并思 范例二:J=计算机应用与软件 AND (U=C++ OR U=Basic) NOT M=Visual
作 者:刘军[1]
机构地区:[1]南京工业大学电子与信息工程学院,江苏南京210009
出 处:《河海大学学报(自然科学版)》2013年第2期177-183,共7页Journal of Hohai University(Natural Sciences)
基 金:国家自然科学基金(60673185);教育部留学回国人员科研启动基金(200711108)
摘 要:针对非相容表构建决策树中属性约简计算复杂度较高的问题,基于粗糙集分辨关系和粒计算理论,建立一种对相容表和非相容表通用的构建决策树算法。该算法首先根据非相容表的逻辑关系式证明其可经过变量替代等价于相容表;然后根据等价表的特点,以属性粒为基本单位,用属性粒的可分辨量、属性粒相对于决策属性的类别数量、决策表中相同行的数量这3个基本参数确定属性的优劣,并以最优属性为划分属性自上向下直接构建决策树。实例分析结果表明,应用该算法构建决策树具有简洁、有效性。To overcome the difficulties in complex attribute reduction calculation during decision tree construction for incompatible tables,a general algorithm for constructing a decision tree suitable for compatible and incompatible tables is put forward based on the rough set of resolution relations and the granulation computing theory.According to the logical relationship of the incompatible table,this algorithm proves that the table created by the substitution of variables can be equivalent to a compatible table.Then,according to the features of the equivalent table,with the attribute granulation being regarded as the basic unit,three basic parameters,the resolution volume of attribute granulation,the class of attribute granulation relative to the decision attribute,and the number of same lines in the decision table,are used to determine the importance of the attribute and construct the decision tree from top to bottom with the best attribute.Experiments and comparison show that this algorithm for constructing the decision tree is simple and efficient.
分 类 号:TP18[自动化与计算机技术—控制理论与控制工程]
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在链接到云南高校图书馆文献保障联盟下载...
云南高校图书馆联盟文献共享服务平台 版权所有©
您的IP:3.16.90.150