基于SRM自组织多区域覆盖的可拒绝近邻分类算法研究

doi:10.3724/SP.J.1146.2007.01360

摘要
图/表
参考文献
相关文章 (4)

全文: PDF (220 KB)
输出: BibTeX | EndNote (RIS)

摘要该文依据区分与划分相结合的可拒绝模式识别思路，提出了高维空间海量训练样本情况下的基于结构风险最小化决策的自组织多区域多球覆盖可拒绝近邻分类算法。该方法利用同类样本之间相互接近的特性，通过结构风险最小化紧几何覆盖策略，选择训练样本，通过自组织多区域多球覆盖模型构成同类样本的划分性描述，达到拒绝识别非训练类样本的目的，最后通过k近邻相互区分性比较确定真实类别。仿真实验结果表明该文的思路是合理可行的，在实际应用领域具有一定价值。

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	胡正平
	贾千文

关键词 ：可拒绝模式识别, 结构风险最小化原理, 近邻分类

Abstract：According to the mode of rejecting pattern recognition principles, which is based on “matter description” and “matter separation” in uniform statistical pattern recognition, a rejecting nearest neighbor classifier based on structural risk minimization self-organization multiple region covering model is presented in this paper. This new model is better closer to the actual instance, rather than traditional statistical pattern recognition only using “optimal separation”as its main principle. Firstly, the optimal samples are selected from the training samples based on structural risk minimization, which is used for same class pattern matter description. Then the kNN distinguish is as a following step to identified the exact class. The simulation experimental results show that this method is valid and efficient.

Key words： Rejecting pattern recognition Structural risk minimization Nearest neighbor classifier

收稿日期: 2007-08-27

PACS:

TP391.4

基金资助:

河北省自然科学基金(F2008000891)，燕山大学博士基金项目(B287)和北京大学视觉与听觉信息处理国家重点实验室开放基金 (0507)资助课题

引用本文:

胡正平;贾千文. 基于SRM自组织多区域覆盖的可拒绝近邻分类算法研究[J]. 电子与信息学报, 2009, 31(2): 293-296 . Hu Zheng-ping; Jia Qian-wen. Rejecting Nearest Neighbor Classifier Based on Structural Risk Minimization Principle Self-organization Multiple Region Covering Model. , 2009, 31(2): 293-296 .

链接本文:

http://jeit.ie.ac.cn/CN/10.3724/SP.J.1146.2007.01360 或 http://jeit.ie.ac.cn/CN/Y2009/V31/I2/293