oda-hub/literature-to-facts

Extracting facts from publications

literature-to-facts

How to contribute to literature analysis?

prepared-for: scienstists, contributors

simply add a function following the example:

https://github.com/cdcihub/literature-to-facts/blob/master/facts/gcn.py#L134

consider adding a test for your and our safety:

https://github.com/cdcihub/literature-to-facts/blob/master/tests/test_gcn.py

Is this like google, finding keywords?

not really. It is extracting structured data, propositions, encoded in RDF. E.g. GRBXXX is-detected-by Swift/BAT.

It can also extract keywords, but unlike google, it, for the moment, uses fixed set of keywords.

see linked-data concept for mode ideas.

Try it locally

for ATels:

Not all atels are ingested, only "interesting" ones (with some useful attributes).

Currently, there are 836 interesting ATels.

to parse atels from html:

python -m  facts.atel -d parse-html ~/ATels.html

or fetch last ones:

python -m  facts.atel -d fetch

to extract from atels

python -m facts.learn learn -t

to extract from atels

python -m facts.learn learn -t

it will store knowledge.n3

which is then published to the kb (given the permissions)

python -m facts.learn publish