Classification by Pairwise Coupling
Classification by Pairwise Coupling
复制标题
DOI：
10.1214/aos/1028144844
复制
发表时间：
1997-12
期刊：
影响因子：
--
通讯作者：
T. Hastie;R. Tibshirani
中科院分区：
文献类型：
--
作者：
T. Hastie;R. Tibshirani
我们讨论了一种策略，涉及到估计类概率为每对类，然后耦合估计在一起的多分类。耦合模型类似于用于配对比较的Bradley-Terry方法。我们研究了类概率估计的性质，并在模拟数据集中检查该过程的性能。所使用的分类器包括线性判别式和最近邻：支持向量机的应用程序也简要介绍。
We discuss a strategy for polychotomous classification that involves estimating class probabilities for each pair of classes, and then coupling the estimates together. The coupling model is similar to the Bradley-Terry method for paired comparisons. We study the nature of the class probability estimates that arise, and examine the performance of the procedure in simulated datasets. The classifiers used include linear discriminants and nearest neighbors: application to support vector machines is also briefly described.