Applying Weights in the Functioning of the Dynamic Classifier Selection Method
Applying Weights in the Functioning of the Dynamic Classifier Selection Method
复制标题
DOI：
10.1109/sbrn.2006.11
复制
发表时间：
2006-10
期刊：
影响因子：
--
通讯作者：
D. Fagundes;A. Canuto
中科院分区：
文献类型：
--
作者：
D. Fagundes;A. Canuto
在多分类器系统中，有两种主要的方法来联合收割机组合分类器的输出，它们是：基于组合的方法和基于选择的方法。在基于选择的方法中，只需要一个分类器来正确地分类输入模式。分类器的选择通常基于当前决策的确定性。另一方面，权重的使用对于多分类器系统的最终决策非常有用，因为它可以为每个分类器提供置信度。本文提出了一种调查使用两个置信度（权重）的动态分类方法，这是一种基于选择的方法的运作。本文的主要目的是分析在基于选择的方法中使用权重的好处，以及哪一种更适合使用。
There are two main approaches to combine the output of classifiers within a multi-classifier system, which are: combination-based and selection-based methods. In selectionbased methods, only one classifier is needed to correctly classify the input pattern. The choice of a classifier is typically based on the certainty of the current decision. On the other hand, the use of weights can be very useful for the final decision of a multi-classifier system since it can provide a confidence degree for each classifier. This paper presents an investigation of using two confidence measures (weights) in the functioning of the dynamic classifier method, which is a selection-based method. The main aim of this paper is to analyze the benefits of using weights in a selection-based method and which one is more suitable to be used.