Volume 10, September 2006
|Page(s)||340 - 355|
|Published online||08 September 2006|
Nearest neighbor classification in infinite dimension
IRISA /INRIA Campus de Beaulieu
35042 Rennes Cédex, France. Frederic.Cerou@inria.fr
2 Université de Rennes 2, Campus de Villejan, 35043 Rennes Cedex, France. Arnaud.Guyader@uhb.fr
Revised: 24 March 2006
Revised: 6 April 2006
Let X be a random element in a metric space (F,d), and let Y be a random variable with value 0 or 1. Y is called the class, or the label, of X. Let (Xi,Yi)1 ≤ i ≤ n be an observed i.i.d. sample having the same law as (X,Y). The problem of classification is to predict the label of a new random element X. The k-nearest neighbor classifier is the simple following rule: look at the k nearest neighbors of X in the trial sample and choose 0 or 1 for its label according to the majority vote. When , Stone (1977) proved in 1977 the universal consistency of this classifier: its probability of error converges to the Bayes error, whatever the distribution of (X,Y). We show in this paper that this result is no longer valid in general metric spaces. However, if (F,d) is separable and if some regularity condition is assumed, then the k-nearest neighbor classifier is weakly consistent.
Mathematics Subject Classification: 62H30
Key words: Classification / consistency / non parametric statistics.
© EDP Sciences, SMAI, 2006
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.