TY - GEN
T1 - Kernel difference-weighted k-nearest neighbors classification
AU - Zuo, Wangmeng
AU - Wang, Kuanquan
AU - Zhang, Hongzhi
AU - Zhang, Dapeng
PY - 2007/12/1
Y1 - 2007/12/1
N2 - Nearest Neighbor (NN) rule is one of the simplest and most important methods in pattern recognition. In this paper, we propose a kernel difference-weighted k-nearest neighbor method (KDF-WKNN) for pattern classification. The proposed method defines the weighted KNN rule as a constrained optimization problem, and then we propose an efficient solution to compute the weights of different nearest neighbors. Unlike distance-weighted KNN which assigns different weights to the nearest neighbors according to the distance to the unclassified sample, KDF-WKNN weights the nearest neighbors by using both the norm and correlation of the differences between the unclassified sample and its nearest neighbors. Our experimental results indicate that KDF-WKNN is better than the original KNN and distanceweighted KNN, and is comparable to some state-of-the-art methods in terms of classification accuracy.
AB - Nearest Neighbor (NN) rule is one of the simplest and most important methods in pattern recognition. In this paper, we propose a kernel difference-weighted k-nearest neighbor method (KDF-WKNN) for pattern classification. The proposed method defines the weighted KNN rule as a constrained optimization problem, and then we propose an efficient solution to compute the weights of different nearest neighbors. Unlike distance-weighted KNN which assigns different weights to the nearest neighbors according to the distance to the unclassified sample, KDF-WKNN weights the nearest neighbors by using both the norm and correlation of the differences between the unclassified sample and its nearest neighbors. Our experimental results indicate that KDF-WKNN is better than the original KNN and distanceweighted KNN, and is comparable to some state-of-the-art methods in terms of classification accuracy.
KW - K-nearest neighbor
KW - Kernel methods
KW - Pattern classification
UR - http://www.scopus.com/inward/record.url?scp=38049081660&partnerID=8YFLogxK
M3 - Conference article published in proceeding or book
SN - 9783540742012
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 861
EP - 870
BT - Advanced Intelligent Computing Theories and Applications
T2 - 3rd International Conference on Intelligent Computing, ICIC 2007
Y2 - 21 August 2007 through 24 August 2007
ER -