CLEF EMED 2019 Dataset

This is the dataset used in our work for our SALt method (code is available at https://github.com/levnikmyskin/salt). We make here available the tf-idf vectors used in our experiments for the CLEF EMED 2019 dataset and their respective labels.

The file is a zip compressed archive. In order to use it, simply uncompress it. Every topic has a "data" and a "labels" file, saved in npz format. For the "data" objects, use scipy.sparse.load_npz; for the "labels" objects, use numpy.load instead.