You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I love paperscape, and also would like to apply this great idea on PubMed, or maybe PubMed Central (4.8 Million full articles instead of 27.3 abstracts like in PubMed).
What would be in your opinion the best point to start therefor?
The text was updated successfully, but these errors were encountered:
Basically most of the metadata for pubmed articles is available since pubmed allows to download xml files with relevant information (references, etc) for most of the recent papers (they have a tool for batch downloading the files). The articles that do not have references annotated in the xml would have to be manually annotated.
The number of times each article is cited and other things should be calculated quite easily.
I was able to extract that information into csv files which I guess can be loaded into the database.
Now I have been trying to generate a map using all the default data and database schemes but I am having trouble connecting to the MySQL database.
[repost of a comment in the blog]
The text was updated successfully, but these errors were encountered: