- -

On the use of diagonal and class-dependent weighted distances for the Probabilistic k-nearest neighbor

RiuNet: Institutional repository of the Polithecnic University of Valencia

Share/Send to

Cited by

Statistics

On the use of diagonal and class-dependent weighted distances for the Probabilistic k-nearest neighbor

Show full item record

Paredes Palacios, R.; Girolami, M. (2011). On the use of diagonal and class-dependent weighted distances for the Probabilistic k-nearest neighbor. En Pattern Recognition and Image Analysis. Springer Verlag (Germany). 6669:265-272. doi:10.1007/978-3-642-21257-4_33

Por favor, use este identificador para citar o enlazar este ítem: http://hdl.handle.net/10251/36702

Files in this item

Item Metadata

Title: On the use of diagonal and class-dependent weighted distances for the Probabilistic k-nearest neighbor
Author:
UPV Unit: Universitat Politècnica de València. Departamento de Sistemas Informáticos y Computación - Departament de Sistemes Informàtics i Computació
Universitat Politècnica de València. Instituto Universitario Mixto Tecnológico de Informática - Institut Universitari Mixt Tecnològic d'Informàtica
Issued date:
Abstract:
A probabilistic k-nn (PKnn) method was introduced in [13] under the Bayesian point of view. This work showed that posterior inference over the parameter k can be performed in a relatively straightforward manner using Markov ...[+]
Subjects: Statistical Pattern-Recognition , Classification , Metrics , Rules , Error
Copyrigths: Cerrado
ISBN: 978-3-642-21256-7
Source:
Pattern Recognition and Image Analysis. (issn: 0302-9743 )
DOI: 10.1007/978-3-642-21257-4_33
Publisher:
Springer Verlag (Germany)
Publisher version: http://link.springer.com/chapter/10.1007/978-3-642-21257-4_33
Conference name: 5th Iberian Conference, IbPRIA 2011
Conference place: Las Palmas de Gran Canaria, Spain
Conference date: June 8-10, 2011
Series: Lecture Notes in Computer Science;vol. 6669
Type: Capítulo de libro

This item appears in the following Collection(s)

Show full item record