An Adaptive Rule Based on Unknown Pattern for Improving K-Nearest Neighbor Classifier
Authors:
- I-Ling ChenICI-Ling Chen
- Kai-Chih PaiKPKai-Chih Pai
- Bor-Chen KuoBKBor-Chen Kuo
- Cheng-Hsuan LiCLCheng-Hsuan Li
- 1 November 2010
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 331-334
Abstract
One of popular and simple pattern classification algorithms is the k-nearest neighbor rule. However, it often fails to work well when patterns of different classes overlap in some regions in the feature space. To overcome this problem, many researches strive for developing various adaptive or discriminatory metrics to improve its performance for classification, recently. In this paper, we proposed a simple adaptive nearest neighbor rule on distance measure for two objects. First one is to separate the overlapping data, and the second one is to avoid the influence of outliers. From the experimental results, our method is robust for the choice of the number of k and outperforms than k-nearest neighbor classifier.Keywords
This publication has 7 references indexed in Scilit:
- Improving nearest neighbor rule with a simple adaptive distance measurePattern Recognition Letters, 2007
- An Optimization Algorithm of K-NN Classification2006
- Signal Theory Methods in Multispectral Remote Sensing2003
- Locally adaptive metric nearest-neighbor classificationIEEE Transactions on Pattern Analysis and Machine Intelligence, 2002
- A ?soft?K-nearest neighbor voting schemeInternational Journal of Intelligent Systems, 2001
- Toward an Explanatory Similarity Measure for Nearest-Neighbor Classification2000
- Discriminant adaptive nearest neighbor classificationIEEE Transactions on Pattern Analysis and Machine Intelligence, 1996