Evaluating the Ambiguities between Two Classes via Euclidean Distance

Jing-Doo Wang ； Hsiang-Chuan Liu

《Asian Journal of Health and Information Sciences》 4卷1期 (2009/03) Pp. 21-35

https://doi.org/10.6412/AJHIS.200903.0021

並列摘要

The classification was a supervised learning approach and, therefore, the class structure was specified in advance by domain experts. The goal of this paper is to evaluate the degree of ambiguity between two classes in the existing class structure via Euclidean distance. In this paper, Distinguishable Distance Ratio (DDR) and Class Ambiguity Ratio (CAR) between two classes are proposed to indicate the degree of the ambiguity between classes. The degree of class ambiguity between two classes is expected to be high if the value of DDR is low and the value of CAR is high. The experimental resources for class structure evaluation includes ”Iris Plant,” ”Wine Recognition” and ”Glass Identification,” and the values of DDR and CAR were found to reveal the degree of class ambiguity. This work offers domain expertise an approach to examine the fitness of class structure, if necessary. To our knowledge, we are the first to address the problem of the ambiguity of class structure.

並列關鍵字

classification ； taxonomy ； class structure ； class ambiguity