Asian Journal of Information Technology
ISSN: Online 1993-5994ISSN: Print 1682-3915
Abstract
This paper deals with the use of the classification rule by an alternative of the nearest neighbors method. We describe first an algorithm based on the partition of the training space in adjacent hyper-cubic cells, then a classification algorithm with two options: one gives an approximate nearest neighbor, the other an exact nearest neighbor. Finally, in a third part, we give results which were obtained by the application on real data. to illustrate the advantage of this method.
How to cite this article:
L. Zouaoui , H. Akdag , M. Bedda and M. Boughazi . Classification by an Alternative of the Nearest Neighbor Method.
DOI: https://doi.org/10.36478/ajit.2004.907.912
URL: https://www.makhillpublications.co/view-article/1682-3915/ajit.2004.907.912