PASCAL - Pattern Analysis, Statistical Modelling and Computational Learning

Similarity Learning for Nearest Neighbor Classification
Ali Mustafa Qamar, Eric Gaussier, Jean-Pierre Chevallet and Joo Hwee Lim
In: Eighth IEEE International Conference on Data Mining 2008, 15-19 Dec 2008, Pisa, Italy.


In this paper, we propose an algorithm for learning a general class of similarity measures for kNN classification. This class encompasses, among others, the standard cosine measure, as well as the Dice and Jaccard coefficients. The algorithm we propose is an extension of the voted perceptron algorithm and allows one to learn different types of similarity functions (either based on diagonal, symmetric or asymmetric similarity matrices). The results we obtained show that learning similarity measures yields significant improvements on several collections, for two prediction rules: the standard kNN rule, which was our primary goal, and a symmetric version of it.

EPrint Type:Conference or Workshop Item (Oral)
Project Keyword:Project Keyword UNSPECIFIED
Subjects:Theory & Algorithms
ID Code:4385
Deposited By:Ali QAMAR
Deposited On:31 March 2009