TU Darmstadt / ULB / TUbiblio

On Minimizing the Position Error in Label Ranking

Hüllermeier, Eyke and Fürnkranz, Johannes (2007):
On Minimizing the Position Error in Label Ranking.
[Online-Edition: http://www.ke.informatik.tu-darmstadt.de/publications/report...],
[Report]

Abstract

Conventional classification learning allows a classifier to make a one shot decision in order to identify the correct label. However, in many practical applications, the problem is not to give a single estimation, but to make repeated suggestions until the correct target label has been identified. Thus, the learner has to deliver a label ranking, that is, a ranking of all possible alternatives. In this paper, we discuss a loss function, called the position error, which is suitable for evaluating the performance of a label ranking algorithm in this setting. Moreover, we propose “ranking through iterated choice”, a general strategy for extending any multi-class classifier to this scenario. Its basic idea is to reduce label ranking to standard classification by successively predicting a most likely class label and retraining a model on the remaining classes. We demonstrate empirically that this procedure does indeed reduce the position error in comparison with a conventional approach that ranks the classes according to their estimated probabilities. Besides, we also address the issue of implementing ranking through iterated choice in a computationally efficient way.

Item Type: Report
Erschienen: 2007
Creators: Hüllermeier, Eyke and Fürnkranz, Johannes
Title: On Minimizing the Position Error in Label Ranking
Language: English
Abstract:

Conventional classification learning allows a classifier to make a one shot decision in order to identify the correct label. However, in many practical applications, the problem is not to give a single estimation, but to make repeated suggestions until the correct target label has been identified. Thus, the learner has to deliver a label ranking, that is, a ranking of all possible alternatives. In this paper, we discuss a loss function, called the position error, which is suitable for evaluating the performance of a label ranking algorithm in this setting. Moreover, we propose “ranking through iterated choice”, a general strategy for extending any multi-class classifier to this scenario. Its basic idea is to reduce label ranking to standard classification by successively predicting a most likely class label and retraining a model on the remaining classes. We demonstrate empirically that this procedure does indeed reduce the position error in comparison with a conventional approach that ranks the classes according to their estimated probabilities. Besides, we also address the issue of implementing ranking through iterated choice in a computationally efficient way.

Divisions: 20 Department of Computer Science
20 Department of Computer Science > Knowl­edge En­gi­neer­ing
Date Deposited: 24 Jun 2011 15:25
Official URL: http://www.ke.informatik.tu-darmstadt.de/publications/report...
Identification Number: TUD-KE-2007-04
Export:

Optionen (nur für Redakteure)

View Item View Item