Nearest Neighbor Classification Using Cam Weighted Distance
Nearest Neighbor (NN) classification assumes class conditional probabilities to be locally constant, and suffers from bias in high dimensions with a small sample set. In this paper, we propose a novel cam weighted distance to ameliorate the curse of dimensionality. Different from the existing neighbor-based methods, which only analyze a small space emanating from the query sample, the proposed nearest neighbor classification using cam weighted distance (CamNN) optimizes the distance measure based on the analysis of the inter-prototype relationships. Experiments show that CamNN significantly outperforms one nearest neighbor classification (1-NN) and k-nearest neighbor classification (k-NN) in most benchmarks, while its computational complexity is competitive with 1-NN classification.
Unable to display preview. Download preview PDF.
- 6.Friedman, J.H.: Flexible metric nearest neighbor classication. The Pennsylvania State University CiteSeer Archives. (September 24 1999)Google Scholar