ON THE RELATION OF PERFORMANCE TO EDITING IN NEAREST NEIGHBOR RULES
作者： Jack Koplowitz , Thomas A. Brown
DOI: 10.1016/0031-3203(81)90102-3
关键词:
摘要: A general class of editing schemes is examined which allows for the relabeling as well deletion samples. It shown that there a trade-offbetween asymptotic performance and sample can adversely affect finite performance. kk' rule proposed to minimize proportion deleted slight modification introduced an exact analysis in any dimension. Nearest neighbors Editing Probability misclassification NN rules Asymptotic
elsevier.com
sci-hub.se 参考文章(6)
C S Penrod, T J Wagner, Another Look at the Edited Nearest Neighbor Rule. ,(1976)
Ivan Tomek, A Generalization of the k-NN Rule IEEE Transactions on Systems, Man, and Cybernetics. ,vol. SMC-6, pp. 121- 126 ,(1976) , 10.1109/TSMC.1976.5409182
Dennis L. Wilson, Asymptotic Properties of Nearest Neighbor Rules Using Edited Data systems man and cybernetics. ,vol. 2, pp. 408- 421 ,(1972) , 10.1109/TSMC.1972.4309137
T. Cover, P. Hart, Nearest neighbor pattern classification IEEE Transactions on Information Theory. ,vol. 13, pp. 21- 27 ,(1967) , 10.1109/TIT.1967.1053964
T. Wagner, Convergence of the edited nearest neighbor (Corresp.) IEEE Transactions on Information Theory. ,vol. 19, pp. 696- 697 ,(1973) , 10.1109/TIT.1973.1055059
T. Wagner, Convergence of the nearest neighbor rule IEEE Transactions on Information Theory. ,vol. 17, pp. 566- 571 ,(1971) , 10.1109/TIT.1971.1054698