This demo uses sample code from a ContentMine-API dump to Zenodo and can be recreated with this source code.
The raw data consists of a dump from the ContentMine-API based on a query for "Zika" on EuropePMC, which can be downloaded from Zenodo. It has been preprocessed and the features for each visualization have been extracted, see here for the source code.
The data consists of 34039 individual facts extracted from 503 publications in the range of April 1998 to September 2016 (as indicated by first publication date by EuropePMC). Facts are extracted by matching the fulltext with our dictionaries, a collection of controlled vocabularies of words interesting for a domain or a specific research question.
This graph shows the distribution of papers over journals, with absolute counts on the left hand scale, and cumulative percentage on the right hand scale.