Advances in Class Noise Detection
Advances in Class Noise Detection
复制标题
DOI：
10.3233/978-1-60750-606-5-1105
复制
发表时间：
2010-08
期刊：
影响因子：
--
通讯作者：
Borut Sluban;D. Gamberger;N. Lavrač
中科院分区：
文献类型：
--
作者：
Borut Sluban;D. Gamberger;N. Lavrač
为了提高归纳分类器的分类精度，通常在数据预处理中使用噪声滤波。我们的目标不同：我们的目标是检测噪声实例，以便领域专家在数据理解阶段进行检查。因此，我们的噪声检测算法应该具有较高的类噪声检测精度，其中精度和召回率之间的权衡是使用F度量来建模的。已经开发了新的类别噪声检测算法的变体，包括高一致性随机森林过滤器，它确保了识别错误数据实例的非常高的精度。
Noise filtering is usually used in data preprocessing to improve the accuracy of induced classifiers. Our goal is different: we aim at detecting noisy instances to be inspected by the domain expert in the phase of data understanding. Consequently, our noise detection algorithms should have high precision of class noise detection, where the precision-recall trade-off is modeled using the F-measure. New variants of class noise detection algorithms have been developed, including the high agreement random forest filter which ensures very high precision of identified erroneous data instances.