TU Darmstadt / ULB / TUbiblio

Meta-Learning Rule Learning Heuristics

Janssen, Frederik and Fürnkranz, Johannes (2007):
Meta-Learning Rule Learning Heuristics.
[Online-Edition: http://www.ke.informatik.tu-darmstadt.de/publications/report...],
[Report]

Abstract

The goal of this paper is to investigate to what extent a rule learning heuristic can be learned from experience. Our basic approach is to learn a large number of rules and record their performance on the test set. Subsequently, we train regression algorithms on predicting the test set performance from training set characteristics. We investigate several variations of this basic scenario, including the question whether it is better to predict the performance of the candidate rule itself of the resulting final rule. Our experiments on a number of independent evaluation sets show that the learned heuristics outperform standard rule learning heuristics. We also analyze their behavior in coverage space.

Item Type: Report
Erschienen: 2007
Creators: Janssen, Frederik and Fürnkranz, Johannes
Title: Meta-Learning Rule Learning Heuristics
Language: English
Abstract:

The goal of this paper is to investigate to what extent a rule learning heuristic can be learned from experience. Our basic approach is to learn a large number of rules and record their performance on the test set. Subsequently, we train regression algorithms on predicting the test set performance from training set characteristics. We investigate several variations of this basic scenario, including the question whether it is better to predict the performance of the candidate rule itself of the resulting final rule. Our experiments on a number of independent evaluation sets show that the learned heuristics outperform standard rule learning heuristics. We also analyze their behavior in coverage space.

Divisions: 20 Department of Computer Science
20 Department of Computer Science > Knowl­edge En­gi­neer­ing
Date Deposited: 24 Jun 2011 15:26
Official URL: http://www.ke.informatik.tu-darmstadt.de/publications/report...
Identification Number: TUD-KE-2007-02
Export:

Optionen (nur für Redakteure)

View Item View Item