Decision templates for multiple classifier fusion: an experimental comparison
Decision templates for multiple classifier fusion: an experimental comparison
复制标题
DOI：
10.1016/s0031-3203(99)00223-x
复制
发表时间：
2001-02-01
影响因子：
8
通讯作者：
Duin, RPW
中科院分区：
文献类型：
--
作者：
Kuncheva, LI;Bezdek, JC;Duin, RPW
多分类器融合可以生成比每个组成分类器更准确的分类。融合通常基于固定的组合规则，如乘积和平均值。只有在严格的概率条件下，这些规则才是合理的。我们在这里提出一个简单的规则，使类合并器适应应用程序。C决策模板（每类一个）用用于分类器集合的相同训练集合来估计。然后，通过某种相似性度量将这些模板与新传入对象的决策配置文件进行匹配。我们比较了我们的模型的11个版本与其他14种技术的分类器融合的Satimage和音素数据集从数据库ELENA。我们的研究结果表明，决策模板的基础上，积分类型的相似性措施是优于其他计划的两个数据集上的上级。(C)2000模式识别学会。由Elsevier Science Ltd.出版，版权所有。
Multiple classifier fusion may generate more accurate classification than each of the constituent classifiers. Fusion is often based on fixed combination rules like the product and average. Only under strict probabilistic conditions can these rules be justified. We present here a simple rule for adapting the class combiner to the application. c decision templates (one per class) are estimated with the same training set that is used for the set of classifiers. These templates are then matched to the decision profile of new incoming objects by some similarity measure. We compare 11 versions of our model with 14 other techniques for classifier fusion on the Satimage and Phoneme datasets from the database ELENA. Our results show that decision templates based on integral type measures of similarity are superior to the other schemes on both data sets. (C) 2000 Pattern Recognition Society. Published by Elsevier Science Ltd. All rights reserved.