Pattern recognition via projection – based k – NN rules
Date
2008-06
Authors
Fraiman, Ricardo
Justel, Ana
Svarc, Marcela
relationships.isContributorOfPublication
Journal Title
Journal ISSN
Volume Title
Publisher
Universidad de San Andrés. Departamento de Matemáticas y Ciencias
Abstract
We introduce a new procedure for pattern recognition, based on the concepts of random projections and nearest neighbors. It can be thought as an improvement of the classical nearest neighbors classification rules. Besides the concept of neighbors we introduce the notion of district, a larger set which will be projected. Then we apply one dimensional k-NN methods to the projected data on randomly selected directions. In this way we are able to provide a method with some robustness properties and more accurate to handle high dimensional data. The procedure is also universally consistent. We challenge the method with the Isolet data where we obtain a very high classification score.
Description
Fil: Fraiman, Ricardo. Universidad de San Andrés. Departamento de Matemática y Ciencias; Argentina.
Fil: Justel, Ana. Universidad de San Andrés. Departamento de Matemática y Ciencias; Argentina.
Fil: Svarc, Marcela. Universidad de San Andrés. Departamento de Matemática y Ciencias; Argentina.
Fil: Justel, Ana. Universidad de San Andrés. Departamento de Matemática y Ciencias; Argentina.
Fil: Svarc, Marcela. Universidad de San Andrés. Departamento de Matemática y Ciencias; Argentina.
Keywords
Multivariate analysis , Robust statistics , Pattern perception