排序方式： 共有1条查询结果，搜索用时 0 毫秒
1
1.
A non-parametric method for supervised pattern recognition is presented. The method is of the classmodelling type, meaning that a classification rule is developed for each class, using the dissimilaritiesbetween the objects of the class. The dissimilarities between the objects within a class are related to thedistances between all pairs of training objects. As distance metric, a measure is proposed that takes thecorrelation between the interval-scale variables into account, and that moreover can be used for mixedtypes of variables. The classification rule is based on the construction of a boundary in the measurementspace. For the determination of the class boundary, several strategies are proposed and compared. The performance of the technique is evaluated on the basis of several data sets. Comparison with theclass modelling technique UNEQ shows its usefulness for practical applications. 相似文献
1