Answers are grounded in this paper and cite their sources.
The finite-sample risk of the k nearest neighbor classifier that uses a weighted Lᵖ-metric as a measure of class similarity is examined. For a family of classification problems with smooth distributions in mathbbRⁿ, an asymptotic expansion for the risk is obtained in decreasing fractional powers of the reference sample size. An analysis of the leading expansion coefficients reveals that the optimal weighted Lᵖ-metric, that is, the metric that minimizes the finite-sample risk, tends to a weighted Euclidean (i.e., L²) metric as the sample size is increased. Numerical simulations corroborate this finding for a pattern recognition problem with normal class-conditional densities.
No takes yet. Share an insight, caveat, or question.
Snapp et al. (1998) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: