Adaptive voting rules for k-nearest neighbors classifiers

Authors:
R. Rovatti;R. Ragazzoni;Zs. M. Kovàcs;R. Guerrieri
Affiliations:
Department of Electronics, University of Bologna, Italy.;Department of Electronics, University of Bologna, Italy.;Department of Electronics, University of Bologna, Italy.;Department of Electronics, University of Bologna, Italy.
Venue:
Neural Computation
Year:
1995

Citing 8
Cited 3

Recognizing hand-printed letters and digits using backpropagation learning

Neural Computation
Bayesian interpolation

Neural Computation
Local learning algorithms

Neural Computation
Local algorithms for pattern recognition and dependencies estimation

Neural Computation
The evidence framework applied to classification networks

Neural Computation
Review:

Neural Computation
Generalization effects of k-neighbor interpolation training

Neural Computation
Handwritten digit recognition using k nearest-neighbor, radial-basis function, and backpropagation neural networks

Neural Computation

Multiple Classifier Combination for Character Recognition: Revisiting the Majority Voting System and Its Variations

DAS '02 Proceedings of the 5th International Workshop on Document Analysis Systems V
Classified Vector Quantisation and population decoding for pattern recognition

International Journal of Artificial Intelligence and Soft Computing
Combining case-based and similarity-based product recommendation

ECCBR'06 Proceedings of the 8th European conference on Advances in Case-Based Reasoning

Quantified Score

Hi-index	0.00

Visualization

Abstract

A simple form of cooperation between the k-nearest neighbors (NN) approach to classification and the neural-like property of adaptation is explored. A tunable, high level k-nearest neighbors decision rule is defined that comprehends most previous generalizations of the common majority rule. A learning procedure is developed that applies to this rule and exploits those statistical features that can be induced from the training set. The overall approach is tested on a problem of handwritten character recognition. Experiments show that adaptivity in the decision rule may improve the recognition and rejection capability of standard k-NN classifiers.