Feature Selection for Symbolic Data Classification

Feature Selection for Symbolic Data Classification
复制标题

符号数据分类的特征选择

DOI:
10.1007/978-3-642-51175-2_48
复制
发表时间:
1994
期刊:
--
影响因子:
--
通讯作者:
M. Ichino
M. Ichino
中科院分区:
--
文献类型:
--
作者:
M. Ichino

文献摘要

被引文献

相似文献

本文提出了一种处理符号数据的数学模型--笛卡尔空间模型(CSM)。然后,作为Watanabe的丑小鸭定理的类似定理,我们提出了基于CSM上定义的互邻域图(MNG)的伪简单性定理。我们的特征选择方法是在MNG方面实现的。我们提出了一个奇偶校验问题,以说明我们的特征选择方法的有效性。
This paper presents the Cartesian space model (CSM) which is a mathematical model to treat symbolic data. Then, as a similar theorem to theTheorem of the ugly ducklingby Watanabe, we present thePretended simplicity theorembased on themutual neighborhood graph (MNG) defined on theCSM. Our feature selection method is realized in terms of theMNG. We present a parity problem in order to illustrate the effectiveness of our feature selection method.