Improving Nearest Neighbor Rule with a Simple Adaptive Distance Measure
The k-nearest neighbor rule is one of the simplest and most attractive pattern classification algorithms. However, it faces serious challenges when patterns of different classes overlap in some regions in the feature space. In the past, many researchers developed various adaptive or discriminant metrics to improve its performance. In this paper, we demonstrate that an extremely simple adaptive distance measure significantly improves the performance of the k-nearest neighbor rule.
KeywordsDistance Measure Class Label Lower Error Rate Manhattan Distance Query Pattern
Unable to display preview. Download preview PDF.
- 1.Fix, E., Hodges, J.: Discriminatory analysis, nonparametric discrimination: consistency properties. Tech. Report 4, USAF School of Aviation Medicine. Randolph Field, Texas (1951)Google Scholar
- 2.Friedman, J.: Flexible metric nearest neighbor classification. Technical Report 113, Stanford University Statistics Department (1994)Google Scholar
- 5.Wang, J., Neskovic, P., Cooper, L.N.: An adaptive nearest neighbor rule for classification. In: Yeung, D.S., Liu, Z.-Q., Wang, X.-Z., Yan, H. (eds.) ICMLC 2005. LNCS (LNAI), vol. 3930, Springer, Heidelberg (2006)Google Scholar