4948

Landmarking for Meta-Learning using RapidMiner

Sarah Daniel Abdelmessih, Faisal Shafait, Matthias Reif, Markus Goldstein

RapidMiner Community Meeting and Conference RapidMiner Community Meeting and Conference (RCOMM-10), September 13-16, Dortmund, Germany , Online , 2010
In machine learning, picking the optimal classifier for a given problem is a challenging task. A recent research field called meta-learning automates this procedure by using a meta-classifier in order to predict the best classifier for a given dataset. Using regression techniques, even a ranking of preferred learning algorithms can be determined. However, all methods are based on a prior extraction of meta-features from datasets. Landmarking is a recent method of computing meta-features, which uses the accuracies of some simple classifiers as characteristics of a dataset. Considered as the first meta-learning step in RapidMiner, a new operator called landmarking has been developed. Evaluations based on 90 datasets, mainly from the UCI repository, show that the landmarking features from the proposed operator are useful for predicting classifiers' accuracies based on regression.

Show BibTex:

@inproceedings {
       abstract = {In machine learning, picking the optimal classifier
for a given problem is a challenging task. A recent research
field called meta-learning automates this procedure by using
a meta-classifier in order to predict the best classifier for a
given dataset. Using regression techniques, even a ranking of
preferred learning algorithms can be determined. However, all
methods are based on a prior extraction of meta-features from
datasets. Landmarking is a recent method of computing meta-features, which uses the accuracies of some simple classifiers as
characteristics of a dataset. Considered as the first meta-learning
step in RapidMiner, a new operator called landmarking has been
developed. Evaluations based on 90 datasets, mainly from the UCI
repository, show that the landmarking features from the proposed
operator are useful for predicting classifiers' accuracies based on
regression.},
       number = {}, 
       month = {9}, 
       year = {2010}, 
       title = {Landmarking for Meta-Learning using RapidMiner}, 
       journal = {}, 
       volume = {}, 
       pages = {}, 
       publisher = {Online}, 
       author = {Sarah Daniel Abdelmessih, Faisal Shafait, Matthias Reif, Markus Goldstein}, 
       keywords = {},
       url = {http://www.dfki.de/web/forschung/publikationen/renameFileForDownload?filename=Sarah-Landmarking-RCOMM10.pdf&file_id=uploads_781}
}