Symbolic Pattern Classifiers Based on the Cartesian System Model
Symbolic Pattern Classifiers Based on the Cartesian System Model
复制标题
基于笛卡尔系统模型的符号模式分类器
DOI：
10.1007/978-4-431-65950-1_40
复制
发表时间：
1998
期刊：
影响因子：
--
通讯作者：
H. Yaguchi
中科院分区：
文献类型：
--
作者：
M. Ichino;H. Yaguchi
作为符号模式分类器，本文提出了基于笛卡尔系统模型的面向区域的符号模式分类方法。我们的面向区域的方法能够使用本地有效的信息来区分模式类。这一事实可以实现，至少表面上，一个完美的歧视模式类下的有限的设计集。因此，我们必须在阶级之间的相似性和阶级描述的一般性之间取得平衡。我们从理论上和实验上描述了这一观点，以断言特征选择的重要性，这在任何模式分类问题中都是至关重要的。我们还提出了一个基于符号数据的例子，以说明我们的方法的实用性。
As symbolic pattern classifiers, this paper presents region oriented methods based on theCartesian system modelwhich is a mathematical model to treatsymbolic data. Our region oriented methods are able to use locally effective information to discriminate between pattern classes. This fact may achieve, at least superficially, a perfect discrimination of the pattern classes under a finite design set. Therefore, we have to take a ballance between theseparabilitybetween classes and thegeneralityof class desciptions. We describe this viewpoint theoretically and experimentally in order to assert the importance offeature selectionwhich is essentially important in any pattern classification problem. We present also an example based onsymbolic datain order to illustrate the usefulness of our approach.