EN FR
EN FR


Section: New Results

Resampling and learning

Participant : Alain Célisse.

This aim is to study the k nearest neighbors algorithm in binary classification in two different cases: Passive and active learning. The choice of k is addressed by cross-validation (resampling). In particular, we try to discover the influence of the cutting parameter on which depends the cross-validation with the retained k value.

It is a joined work with Tristan Mary-Huard [16] .