Title :
A wavelet-based filtering approach to functional bipartite ranking
Author :
Clémençon, S. ; Depecker, M.
Author_Institution :
Dept. TSI, Telecom ParisTech, Paris, France
Abstract :
It is the purpose of this paper to investigate the bipartite ranking task from the perspective of functional data analysis (FDA). Precisely, given a collection of independent copies of a (possibly sampled) random curve X = (X(t))tϵ[0,1] taking its values in a function space X, with a locally smooth autocorrelation structure and to which a binary label Y ϵ {-1, +1} is randomly assigned, the goal is to learn a scoring functions: X → R with optimal ROC curve. Based on nonlinear wavelet-based approximation, it is shown how to select compact finite dimensional representations of the input curves in order to build accurate ranking rules, using recent advances in the ranking problem for multivariate data with binary feedback.
Keywords :
approximation theory; filtering theory; functional analysis; wavelet transforms; FDA; ROC curve; binary feedback; bipartite ranking task; functional bipartite ranking; functional data analysis; input curves; multivariate data; nonlinear wavelet-based approximation; wavelet-based filtering approach; Approximation algorithms; Approximation methods; Filtering; Signal processing; Signal processing algorithms; Stochastic processes; Wavelet transforms; AUC maximization; ROC optimization; bipartite ranking; filtering methods; functional data analysis; supervised learning; wavelet analysis;
Conference_Titel :
Statistical Signal Processing Workshop (SSP), 2011 IEEE
Conference_Location :
Nice
Print_ISBN :
978-1-4577-0569-4
DOI :
10.1109/SSP.2011.5967819