期刊
PATTERN RECOGNITION LETTERS
卷 27, 期 7, 页码 830-837出版社
ELSEVIER SCIENCE BV
DOI: 10.1016/j.patrec.2005.12.001
关键词
statistical pattern recognition; Naive Bayes classifier (NB); optimality of NB; dependent binary features
While Naive Bayes classifier (NB) is Bayes-optimal for independent features, we prove that it is also optimal for two equiprobable classes and two features with equal class-conditional covariances. Although strict optimality does not extend for three features, equal covariances are expected to be beneficial in higher-dimensional spaces. (c) 2005 Elsevier B.V. All rights reserved.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据