Semi-Supervised Weighted Distance Metric Learning for kNN Classification

摘要：

K-Nearest Neighbor (kNN) classification is one of the mast popular machine learning techniques, but it often fails (o work well due to less known information or inappropriate choice of distance melric or the presence of a lot of unrelated features. To handle those issues, we introduce a semi-supervised weighted distance metric learning method for kNN classification. This method uses a graph-based semi-supervised Label Propagation algorithm lo gain more classification information with liny initial classification information, then resorts to improved weighted Relevant Component Analysis to learn a Mahalanobis distance metric, and finally uses learned Mahalanuhis distance metric to replace the original Euclidean distance of kNN classifier. Experiments on UCI dalasets show the effectiveness of our method.

关键词： It nearest neighbor classification semi-superised learning relevant component analysis metric learning

作者: Fangming Gu Dayou Liu Xinying Wang

作者单位: College of Computer Science and Technology Jilin University Changchun,China College of Computer Science and Engineering Changchun University of Technology Changchun. China

会议类型: 国际会议

会议名称: 2010 International Conference on Computer,Mechatronics,Control and Electronic Engineering(2010计算机、机电、控制与电子工程国际会议 CMCE 2010)

会议地点: 长春

会议语种:英文

页码: 406-409

在线出版日期: 2010-08-24（万方平台首次上网日期，不代表论文的发表时间）