首页 | 本学科首页   官方微博 | 高级检索  
     

基于可传递信度模型的k-NN分类规则
引用本文:刘邱云,吴根秀,付雪峰. 基于可传递信度模型的k-NN分类规则[J]. 江西师范大学学报(自然科学版), 2004, 28(3): 221-223
作者姓名:刘邱云  吴根秀  付雪峰
作者单位:江西师范大学,数学与信息科学学院,江西,南昌,330027;江西师范大学,计算机信息工程学院,江西,南昌,330027
基金项目:江西省自然科学基金资助项目(0311041).
摘    要:针对训练模式所属类不确定情形,提出了基于可传递信度模型(TBM)的k-NN分类规则,并结合模糊集理论及可能性理论进行了拓广,最后通过计算机模拟实验将两者作了比较。

关 键 词:k-NN分类规则  TBM  pignistic概率  隶属度  可能性测度
文章编号:1000-5862(2004)03-0221-03

A k-Nearest Neighbor Classification Rule Based on the Transferable Belief Model
LIU Qiu-yun,WU Gen-xiu,FU Xue-feng. A k-Nearest Neighbor Classification Rule Based on the Transferable Belief Model[J]. Journal of Jiangxi Normal University (Natural Sciences Edition), 2004, 28(3): 221-223
Authors:LIU Qiu-yun  WU Gen-xiu  FU Xue-feng
Affiliation:LIU Qiu-yun~1,WU Gen-xiu~1,FU Xue-feng~2
Abstract:Regarding to uncertainty or imprecision of the class membership of some training patterns,this paper presents a k-NN classification rule based on the transferable belief model.And the rule is generalized by combining with fuzzy sets theory and possibility theory.Finally,a computer simulation is performed experimentally to compare it with the generalization.
Keywords:k-NN classification rule  transferable belief model  pignistic probability  membership degree  possibility measure
本文献已被 CNKI 维普 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号