arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1604.01686cs.LG
收藏

一类最近邻变量之间的关系及其精确系综的构造

Relationship between Variants of One-Class Nearest Neighbours and Creating their Accurate Ensembles

University of Toronto(多伦多大学);College of Information Technology, United Arab Emirates University(阿联酋大学信息技术学院)

Shehroz S. Khan, Amir Ahmad

摘要:

在一类分类问题中,只有目标类的数据可用,而非目标类的数据可能完全不存在。在本文中,我们研究一类最近邻(OCNN)分类器及其不同的变体。我们提出了一个理论分析,以显示OCNN的不同变体之间的关系,这些变体可以使用不同的邻居或阈值来识别非目标类的未见过的示例。我们还提出了一种基于四分位数间距的方法,用于在训练过程中没有非目标数据的情况下优化OCNN中使用的参数。然后,我们提出了基于随机子空间和随机投影方法的两种集成方法来创建精确的OCNN集成。我们在15个基准和真实世界的特定领域数据集上测试了所提出的方法,并表明OCNN的随机投影集成性能最好。

英文摘要:

In one-class classification problems, only the data for the target class is available, whereas the data for the non-target class may be completely absent. In this paper, we study one-class nearest neighbour (OCNN) classifiers and their different variants. We present a theoretical analysis to show the relationships among different variants of OCNN that may use different neighbours or thresholds to identify unseen examples of the non-target class. We also present a method based on inter-quartile range for optimising parameters used in OCNN in the absence of non-target data during training. Then, we propose two ensemble approaches based on random subspace and random projection methods to create accurate OCNN ensembles. We tested the proposed methods on 15 benchmark and real world domain-specific datasets and show that random-projection ensembles of OCNN perform best.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑