An Empirical Comparison of Learning Algorithms for Nonparametric Scoring

Marine Depecker (1) , Stéphan Clémençon (2) , Nicolas Vayatis (3)

1 LIST (CEA) - Laboratoire d'Intégration des Systèmes et des Technologies
2 LTCI - Laboratoire Traitement et Communication de l'Information
3 CMLA - Centre de Mathématiques et de Leurs Applications

The TreeRank algorithm was recently proposed as a scoring-based method based on recursive partitioning of the input space. This tree induction algorithm builds orderings by recursively optimizing the Receiver Operating Characteristic (ROC) curve through a one-step optimization procedure called LeafRank. One of the aim of this paper is the indepth analysis of the empirical performance of the variants of TreeRank/LeafRank method. Numerical experiments based on both artificial and real data sets are provided. Further experiments using resampling and randomization, in the spirit of bagging and random forests are developed and we show how they increase both stability and accuracy in bipartite ranking. Moreover, an empirical comparison with other efficient scoring algorithms such as RankBoost and RankSVM is presented on UCI benchmark data sets.

Mots clés: Scoring rules; Ranking trees; ROC curve; AUC maximization; Resampling; Feature Randomization

Domaines: Machine Learning [stat.ML]; Applications [stat.AP]; Méthodologie [stat.ME]

Marine Depecker, Stéphan Clémençon, Nicolas Vayatis. An Empirical Comparison of Learning Algorithms for Nonparametric Scoring. 2011. ⟨hal-00730434⟩