Adding similarity measures based on list and count of publications per source_id

[ste210]

This should be easy and does not require additional download of data

[Michael-E-Rose]

Duplicate to #2, no?

[ste210]

No, I think this is even more simple than #2. Simply looking at the actual publications of the authors, not at their references, and computing the probability that the journal is the same. Also, not urgent

[Michael-E-Rose]

But the other issue is about both cited and citing journals? This seems the other journal is more inclusive, no?

[Michael-E-Rose]

It's the number of publications in the same journal. This needs to be implemented in inform_matches(). Ideally, we construct it in such a way that is interpreted as "The number of publications in sources the original scientist publishes in up until the comparison year."