THE RESEARCH ON AN ADAPTIVE K-NEAREST NEIGHBORS CLASSIFIER

摘要：

k-Nearest neighbor (KNNC) classifier is the most popular non-parametric classifier. But it requires much classification time to search k nearest neighbors of an unlabelled object point, which badly affects its efficiency and performance. In this paper, an adaptive k-nearest neighbors classifier (AKNNC)is proposed. The algorithm can find k nearest neighbors of the unlabelled point in a small hypersphere in order to improve the efficiencies and classify the point. The hyperspheres size can be automatically determined. It requires a quite moderate preprocessing effort, and the cost to classify an unlabelled point is O(ad)+O(k)(1 ≤ a＜＜n). Our experiment shows the algorithm performance is superior to other known algorithms.

关键词： Nearest neighbor pattern recognition hypersphere classification

作者: XIAO-GAO YU XIAO-PENG YU

作者单位: Hubei University of Economics, Wuhan, 430205 Computer School, Wuhan University, Wuhan, 430072

会议类型: 国际会议

会议名称: 2006 International Conference on Machine Learning and Cybernetics(IEEE第五届机器学习与控制论坛)

会议地点: 大连

会议语种:英文

页码: 1241-1246

在线出版日期: 2006-08-13（万方平台首次上网日期，不代表论文的发表时间）