层次分类方法综述 - 模式识别与人工智能.pdf

第26 卷摇 第 12 期摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 模式识别与人工智能摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 Vol.26摇 No.12 摇 2013年12月摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 PR AI摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 Dec.摇 摇2013 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 层 次 分 类 方 法 综 述* 陆彦婷摇 摇 陆建峰摇 摇 杨静宇 (南京理工大学 计算机科学与工程学院摇 南京210094) 摘摇 要摇 层次分类方法利用类别层次结构来分解问题和组织分类器,可有效解决多类分类问题.依据是否要求类 别之间存在显式层次关系,层次分类方法可分为两大类.文中对不要求类别之间存在显式层次关系的层次分类方 法进行综述.首先归纳和阐述此类方法所采用的基本框架,然后介绍和分析其中若干关键技术的研究进展,最后从 算法和应用两个角度对国内外相关研究进行详细叙述,进而对现有方法进行总结,并给出进一步研究的方向. 关键词摇 层次分类,多类分类,类别层次,特征融合,图像分类 中图法分类号摇 TP391.4;O235 A Survey of Hierarchical Classification Methods LU Yan鄄Ting,LUJian鄄Feng,YANGJing鄄Yu (School of Computer Science and Engineering,Nanjing University of Science and Technology, Nanjing 210094) ABSTRACT Hierarchical classification (HC),decomposing problem and organizing the classifiers according to the category hierarchy, is an efficient solution for multi鄄class classification problem. Depending on whether an explicit hierarchical relationship among categories is required,HC methods can be divided into two types. In this paper,the HC methods which do not require explicit hierarchical relationship among categories are reviewed systematically. Firstly,the basic framework of this type of methods is outlined. Then,theresearchprogressesof severalkeytechniquesareelaboratedandanalyzed. Next,the related research work at home and abroad is described in detail from both algorithm and application perspectives. Finally,the existing methods are summarized and several future research directions are pointed out. Key Words摇 Hierarchical Classification, Multi鄄Class Classifi