end{aligned} (16b) where (N_{dk}) is the number of documents assigned to the topic k in time-series data d, (N_{kv}) is the number of documents v assigned to the topic k in all time-series data, and (N_d) is the number of documents included in time-series data d.
These entries allow rich searches to be performed, using syntax such as http://search.datacite.org/ui q=format chemical/x-* which retrieves all deposited instances of documents assigned the media type chemical/x-cml in all repositories that register the metadata with DataCite.org/ui q=format chemical/x-*
To identify medical informatics publications, we searched MEDLINE/PubMed in March 2008 for all documents assigned the term "Medical Informatics" [MeSH terms] or any term in the "Medical Informatics" hierarchy as a Major Topic for each year of the 1987–2006 period.
This MeSH document is simply created by merging the titles and abstracts of a number of documents assigned with that MeSH term.
Classifiers based on only information in the metathesaurus show to perform comparably with a system which models MeSH terms based on a selection of documents assigned to it.
In particular, SwissProt/UniProt collects human polymorphisms and disease mutations (annotated in the HUMSAVAR document) assigned according to literature reports on probable disease association.
The datasets k1a and k1b contain exactly the same set of documents but they differ in how the documents were assigned to different classes.
For the entire study period, approximately 73% of the documents were assigned only one MI-MeSH (Major or Minor Topic) designator.
Sites are unaware of the type of consent document assigned.
The proposed method involves recognising geographical references over the documents and assigning geographical scopes through a graph ranking algorithm.
