TY - GEN
T1 - From Ranking to Classification
T2 - 29th Annual Conference of the German Classification Society - Gesellschaft fur Klassifikation, GfKl 2005
AU - Clémençon, Stéphan
AU - Lugosi, Gábor
AU - Vayatis, Nicolas
N1 - Publisher Copyright:
© 2026, Springer Science and Business Media Deutschland GmbH. All rights reserved.
PY - 2006/1/1
Y1 - 2006/1/1
N2 - In applications related to information retrieval, the goal is not only to build a classifier for deciding whether a document x among a list X is relevant or not, but to learn a scoring function s: X → R for ranking all possible documents with respect to their relevancy. Here we show how the bipartite ranking problem boils down to binary classification with dependent data when accuracy is measured by the AUC criterion. The natural estimate of the risk being of the form of a U-statistic, consistency of methods based on empirical risk minimization is studied using the theory of U-processes. Taking advantage of this specific form, we prove that fast rates of convergence may be achieved under general noise assumptions.
AB - In applications related to information retrieval, the goal is not only to build a classifier for deciding whether a document x among a list X is relevant or not, but to learn a scoring function s: X → R for ranking all possible documents with respect to their relevancy. Here we show how the bipartite ranking problem boils down to binary classification with dependent data when accuracy is measured by the AUC criterion. The natural estimate of the risk being of the form of a U-statistic, consistency of methods based on empirical risk minimization is studied using the theory of U-processes. Taking advantage of this specific form, we prove that fast rates of convergence may be achieved under general noise assumptions.
UR - https://www.scopus.com/pages/publications/105046550398
U2 - 10.1007/3-540-31314-1_25
DO - 10.1007/3-540-31314-1_25
M3 - Conference contribution
AN - SCOPUS:105046550398
SN - 9783540313137
T3 - Studies in Classification, Data Analysis, and Knowledge Organization
SP - 214
EP - 221
BT - From Data and Information Analysis to Knowledge Engineering - Proceedings of the 29th Annual Conference of the Gesellschaft für Klassifikation
A2 - Spiliopoulou, Myra
A2 - Kruse, Rudolf
A2 - Borgelt, Christian
A2 - Nürnberger, Andreas
A2 - Gaul, Wolfgang
PB - Springer Science and Business Media Deutschland GmbH
Y2 - 9 March 2005 through 11 March 2005
ER -