I used au.affiliation_history to get the affiliation history from a list of Authors ID. It worked great but now I am trying to pair the affiliation history with the Year in which the researcher served in the institution that appear in the affiliation history results.
However, I cannot find the way to do this. Is is possible to do this? If so, can you please give me a hint or idea how can I achieve this?
Unfortunately, the information Scopus shares on an author profile on scopus.com is not the same they share via the Author Retrieval API. I think the only way to get to yearly affiliations is to extract them from the publications that you get from the Scopus Search API.
from collections import defaultdict
from pybliometrics.scopus import ScopusSearch
AUTHOR = "7004212771"
q = f"AU-ID({AUTHOR})"
s = ScopusSearch(q)
yearly_affs = defaultdict(lambda: list())
for pub in s.results:
year = pub.coverDate[:4]
auth_idx = pub.author_ids.split(";").index(AUTHOR)
affs = pub.author_afids.split(";")[auth_idx].split("-")
yearly_affs[year].extend(affs)
yearly_affs then contains a list of all affiliations recorded in publications for that year.
Naturally, the list will contain duplicates. If you don't like that, use set() and update() instead.
The .split("-") part for affs is for multiple affiliations (when the researcher reports multiple affiliations on a paper. You might want to use the first reported instead. Then use [0] at the end and append() in the next row.
Also, there will likely be gaps. I recommend turning the yearly_affs into a pandas DataFrame, select for each year the main affiliation, and then fill gaps forward or backward.