Original datasets - reproducibility about bio-cnn HOT 2 CLOSED

Hi SkBlaz,

Here is a direct link to the dataset PMIDs: https://ii.nlm.nih.gov/DataSets/2013_MTI_ML_DataSet.tar

You will need to pull the titles and abstracts for each PMID. For the validation split, I think we simply removed 5000 documents from the training set at random.

If you are interested in MeSH indexing, I recommend taking a look at the BioASQ competition. There are more than 27k MeSH terms, not just 29. The paper you are referring to was a proof of concept on a small number of MeSH terms. I have linked to our paper using all MeSH terms below, as well as the BioASQ competition website:

http://participants-area.bioasq.org/general_information/Task6a/
https://ieeexplore.ieee.org/abstract/document/7349667

from bio-cnn.

Thank you very much! This is exactly what I was looking for.

from bio-cnn.