Iterated Probabilistic Weighted k-Nearest Neighbor Algorithm
J. M. Martínez Otzeta, Basilio Sierra. Abstract:
We report on the application of a new version of the k-NN algorithm named Iterated Probabilistic Weighted k-NN algorithm (IPW-k-NN). This algorithm classifies new cases based on the probability distribution each case has to belong to each class. These probabilities are computed according to the k Nearest Neighbors, as a way of measure the typicalness of a given case with regard to every class.
Keywords:
Supervised Classification, Nearest Neighbor, k-NN, Machine Learning, Pattern Recognition

bibTeX entry:
@inproceedings {otzeta03iterated,
title = "Iterated Probabilistic Weighted k-Nearest Neighbor Algorithm",
author = "J. M. Martínez-Otzeta, Basilio Sierra",
year = "2003",
booktitle = "CAEPIA-TTIA",
volume = "1",
pages = "117--120",
}