Nearest Neighbor Classification
k-nearest neighbor classification; k-NN classification; NN classification
Nearest neighbor classification is a machine learning method that aims at labeling previously unseen query objects while distinguishing two or more destination classes. As any classifier, in general, it requires some training data with given labels and, thus, is an instance of supervised learning. In the simplest variant, the query object inherits the label from the closest sample object in the training set. Common variants extend the decision set from the single nearest neighbor within the training data to the set of k nearest neighbors for any k > 1. The decision rule combines the labels from these kdecision objects, either by simple majority voting or by any distance-based or frequency-based weighting scheme, to decide the predicted label for the query object. Mean-based nearest neighbor classifiers group the training data and work on the means of classes rather than on the individual...
- 1.Ankerst M, Kastenmuller G, Kriegel H-P, Seidl T. Nearest neighbor classification in 3D protein databases. In: Proceedings of the 7th International Conference on Intelligent Systems for Molecular Biology; 1999. p. 34–43.Google Scholar
- 2.Athistos V. Nearest neighbor retrieval and classification. 2007. Available at: http://cs-people.bu.edu/athitsos/nearest-neighbors/
- 5.Efros AA, Berg AC, Mori G, Malik J. Recognizing action at a distance. In: Proceedings of the 9th IEEE Conference on Computer Vision; 2003. p. 726–33.Google Scholar
- 10.Li Y, Yang J, Han J. 1Continuous K-nearest neighbor search for moving objects. In: Proceedings of the 16th International Conference on Scientific and Statistical Database Management; 2004. p. 123–6.Google Scholar
- 11.Shibata T, Kato T, Wada TK-D. Decision tree: an accelerated and memory efficient nearest neighbor classifier. In: Proceedings of the 2003 IEEE International Conference on Data Mining; 2003. p. 641–4.Google Scholar