Clustering Classifiers Learnt from Local Datasets Based on Cosine Similarity
Clustering Classifiers Learnt from Local Datasets Based on Cosine Similarity
复制标题
DOI：
10.1007/978-3-319-25252-0_16
复制
发表时间：
2015-10
影响因子：
5.1
通讯作者：
Kaikai Zhao;Einoshin Suzuki
中科院分区：
文献类型：
--
作者：
Kaikai Zhao;Einoshin Suzuki
本文提出了一种新的度量线性分类器相异度的方法。该方法基于线性分类器超平面的法向量之间的余弦相似性。这种方法的一个显著优点是它具有很好的解释能力，并且需要在数据集之间交换很少的信息。在合成数据集、UCI机器学习库中的数据集和面部表情数据集上的评估表明，该方法在归一化互信息方面优于以前的方法。
In this paper we present a new method to measure the degree of dissimilarity of a pair of linear classifiers. This method is based on the cosine similarity between the normal vectors of the hyperplanes of the linear classifiers. A significant advantage of this method is that it has a good interpretation and requires very little information to exchange among datasets. Evaluations on a synthetic dataset, a dataset from the UCI Machine Learning Repository, and facial expression datasets show that our method outperforms previous methods in terms of the normalized mutual information.