BoostML: An adaptive metric learning for nearest neighbor classification

Date

2010

Authors

Zaidi, N.
Squire, D.
Suter, D.

Editors

Zaki, M.J.
Yu, J.X.
Ravindran, B.
Pudi, V.

Advisors

Journal Title

Journal ISSN

Volume Title

Type:

Conference paper

Citation

Proceedings of the 14th Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD 2010), held in Hyderabad, India, 21-24 June 2010: pp.142-149

Statement of Responsibility

Nayyar Abbas Zaidi, David McG. Squire and David Suter

Conference Name

Pacific-Asia Conference on Knowledge Discovery and Data Mining (14th : 2010 : Hyderabad, India)

Abstract

A Nearest Neighbor (NN) classifier assumes class conditional probabilities to be locally smooth. This assumption is often invalid in high dimensions and significant bias can be introduced when using the nearest neighbor rule. This effect can be mitigated to some extent by using a locally adaptive metric. In this work we propose an adaptive metric learning algorithm that learns an optimal metric at the query point. We learn a distance metric using a feature relevance measure inspired by boosting. The modified metric results in a smooth neighborhood that leads to better classification results. We tested our technique on major UCI machine learning databases and compared the results to state of the art techniques. Our method resulted in significant improvements in the performance of the K-NN classifier and also performed better than other techniques on major databases.

School/Discipline

Dissertation Note

Provenance

Description

Access Status

Rights

Copyright Springer-Verlag Berlin Heidelberg 2010

License

Grant ID

Call number

Persistent link to this record