REUTERS-K500-EX2.arff reference

Waikato / meka

Multi-label classifiers and evaluation procedures using the Weka machine learning framework.

http://waikato.github.io/meka/

GNU General Public License v3.0

200 stars 76 forks source link

REUTERS-K500-EX2.arff reference #40

Closed i02momuj closed 5 years ago

i02momuj commented 7 years ago

Hello,

I would like to know what is the original reference of the REUTERS-K500-EX2.arff dataset, since I would like to use it in one of my experiments.

Thanks!

jmread commented 7 years ago

If I remember correctly, Tsoumakas and Vlahavas describe the process of feature selection on the tmc2007 dataset in their paper "Random k-labelsets: An ensemble method for multilabel classification" , ECML-PKDD, 2007. I applied the same process to one of the Reuters subsets in "Classifier chains for multi-label classification", Mach Learn 2011, to obtain the dataset you mention. The original Reuters data (before preprocessing) is available at http://mulan.sourceforge.net/datasets-mlc.html