Web output
One of the end products of the pipeline is to generate a series of HTML web pages. These are entirely self contained, so they can be viewed locally without the need to upload them to a web server.
The home screen shows a count of papers for each year of the study, along with the number of citations the papers published in that year have recieved over all years. Clicking on a year takes you to the list of publications for that year with selected metadata such as the citation count for that paper.
The keyword metadata is used to build a list of all available keywords.
Clicking on a keyword takes you to a summary page showing all the publications which have that keyword. This page also shows the number of citations each of these papers have, this helps to highlight popular papers about this subject.
The geocoding of the first author is displayed as a global choropleth map.
This page shows some study level metrics which includes a breakdown of numbers of papers published each year, the number of citations received broken down in various ways, and the total values.
This page shows a word cloud generated from the abstract text.
In all cases there is an indication of how many publications are included in the generation of the figure. This is important as not every publication will have all metadata associated with it.