On Partitioning Rules for Bipartite Ranking
نویسندگان
چکیده
The purpose of this paper is to investigate the properties of partitioning scoring rules in the bipartite ranking setup. We focus on ranking rules based on scoring functions. General sufficient conditions for the AUC consistency of scoring functions that are constant on cells of a partition of the feature space are provided. Rate bounds are obtained for cubic histogram scoring rules under mild smoothness assumptions on the regression function. In this setup, it is shown how to penalize the empirical AUC criterion in order to select a scoring rule nearly as good as the one that can be built when the degree of smoothness of the regression function is known.
منابع مشابه
Minimax Learning Rates for Bipartite Ranking and Plug-in Rules
While it is now well-known in the standard binary classification setup, that, under suitable margin assumptions and complexity conditions on the regression function, fast or even super-fast rates (i.e. rates faster than n or even faster than n) can be achieved by plug-in classifiers, no result of this nature has been proved yet in the context of bipartite ranking, though akin to that of classif...
متن کاملRanking forests
The present paper examines how the aggregation and feature randomization principles underlying the algorithm Random Forest (Breiman (2001)) can be adapted to bipartite ranking. The approach taken here is based on nonparametric scoring and ROC curve optimization in the sense of the AUC criterion. In this problem, aggregation is used to increase the performance of scoring rules produced by rankin...
متن کاملRanking Multi-Class Data: Optimality and Pairwise Aggregation
It is the primary purpose of this paper to set the goals of ranking in a multiple-class context rigorously, following in the footsteps of recent results in the bipartite framework. Under specific likelihood ratio monotonicity conditions, optimal solutions for this global learning problem are described in the ordinal situation, i.e. when there exists a natural order on the set of labels. Criteri...
متن کاملA new approach based on data envelopment analysis with double frontiers for ranking the discovered rules from data mining
Data envelopment analysis (DEA) is a relatively new data oriented approach to evaluate performance of a set of peer entities called decision-making units (DMUs) that convert multiple inputs into multiple outputs. Within a relative limited period, DEA has been converted into a strong quantitative and analytical tool to measure and evaluate performance. In an article written by Toloo et al. (2009...
متن کاملConfidence-Weighted Bipartite Ranking
Bipartite ranking is a fundamental machine learning and data mining problem. It commonly concerns the maximization of the AUC metric. Recently, a number of studies have proposed online bipartite ranking algorithms to learn from massive streams of class-imbalanced data. These methods suggest both linear and kernel-based bipartite ranking algorithms based on first and second-order online learning...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2009