Four versions of a k-nearest neighbor algorithm with locally adaptive k are introduced and compared to the basic k-nearest neighbor algorithm (kNN). Locally adaptive kNN algorithms choose the value of k that should be used to classify a query by consulting the results of cross-validation computations in the local neighborhood of the query. Local kNN methods are shown to perform similar to kNN in experiments with twelve commonly used data sets. Encouraging results in three constructed tasks show that local methods can significantly outperform kNN in specific applications. Local methods can be recommended for on-line learning and for applications where different regions of the input space are covered by patterns solving different sub-tasks.
[1]
David Aha.
A study of instance-based algorithms for supervised learning tasks: mathematica:l
,
1990
.
[2]
Belur V. Dasarathy,et al.
Nearest neighbor (NN) norms: NN pattern classification techniques
,
1991
.
[3]
Erkki Oja,et al.
Rival penalized competitive learning for clustering analysis, RBF net, and curve detection
,
1993,
IEEE Trans. Neural Networks.
[4]
Douglas L. Brutlag,et al.
Rapid searches for complex patterns in biological molecules
,
1984,
Nucleic Acids Res..
[5]
Sholom M. Weiss,et al.
Computer Systems That Learn
,
1990
.
[6]
Léon Bottou,et al.
Local Learning Algorithms
,
1992,
Neural Computation.