I just started to have interest into bibliometric analyses, and so far I am struggling to understand the tons of copyright policies. However, these are (from my opinion) quite vague and with too much grey areas.
Let’s say that I constrain my analyses on the metadata from Pubmed (not PMC) considering basic classes in it (i.e., title, publication year, citation number, MeSH terms, author names…);
but I never touch or analyze the abstracts (text mining or anything alike whatsoever on abstracts).
- Do you think that such analyses would pass without copyright infringements?