Welcome to our temporary website describing and hosting the data and user study results used in our current KDD submission. Further below we provide links to the LIBSVM implementation that we used and the PDF documents of our cited related prior work.

The data is primarly contained in different files:

1. Positive samples according to Algorithm 1. PREVIEW / DOWNLOAD
2. Negative samples according to Algorithm 1. PREVIEW / DOWNLOAD
3. Results of the user study, i.e., tasks and their ratings. PREVIEW / DOWNLOAD

For training the classifiers, we have created, as described in the paper, ten smaller training files and one file containing (left-out) test data. DOWNLOAD or see DESCRIPTION.


The LIBSVM library that we used was obtained from https://www.csie.ntu.edu.tw/~cjlin/libsvm/


For ease of access, we provide below the two prior works we reference:

  F. Reinartz, K. Pal, and S. Michel: Mining Entity Rankings. Datenbankspektrum, 2016.

  K. Pal and S. Michel: 2016. A Data Mining Approach to Choosing Categorical Attributes for Ranked Lists. EDBT, Poster Track, 2016.