作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2009, Vol. 35 ›› Issue (16): 183-185. doi: 10.3969/j.issn.1000-3428.2009.16.066

• 人工智能及识别技术 • 上一篇    下一篇

基于TBM的自适应模糊k-NN分类器

刘邱云1,付雪峰2,吴根秀1   

  1. (1. 江西师范大学数学与信息科学学院,南昌 330022;2. 南昌工程学院计算机科学与技术系,南昌 330099)
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2009-08-20 发布日期:2009-08-20

Adaptive Fuzzy k-NN Classifier Based on Transferable Belief Model

LIU Qiu-yun1, FU Xue-feng2, WU Gen-xiu1   

  1. (1. Institute of Mathematics and Informatics, Jiangxi Normal University, Nanchang 330022;2. Department of Computer Science and Technology, Nanchang Institute of Technology, Nanchang 330099)
  • Received:1900-01-01 Revised:1900-01-01 Online:2009-08-20 Published:2009-08-20

摘要: 针对训练模式类标签不精确的识别问题,提出基于可传递信度模型的自适应模糊 k-NN(k-Nearest Neighbor)分类器。利用可传递信度模型结合模糊集理论和可能性理论并运用pignistic变换,对待识别模式真正所属的类做出决策。采用梯度下降最小化误差函数,以实现参数的自适应学习。实验结果表明,该分类器误分类率低、鲁棒性强。

关键词: 可传递信度模型, 自适应, k-NN分类器, pignistic概率, 梯度下降

Abstract: For processing training patterns with imprecise class labels, an adaptive fuzzy k-Nearest Neighbor(k-NN) classifier based on Transferable Belief Model(TBM) is presented. Decision about the true class membership of a pattern can be made to be classified by the combination of TBM, fuzzy sets and possibility theory and the application of the pignistic transformation. The parameters in the classifier are tuned automatically by minimizing error functions through gradient descent. Numerical simulations results show that the classifier has low classification error rates and high robustness.

Key words: Transferable Belief Model(TBM), adaptive, k-Nearest Neighbor(k-NN) classifier, pignistic probability, gradient descent

中图分类号: