first of all thank you very much for this effort in providing the scientific community with a reliable publication explorer .. DBLP has been considered for too much time as the reference without any aim of completeness.
I have noticed that both organization and author ratings suffer some problems due to authors' movements, an aspect quite common in the research field. What I would propose is to make it possible to distinguish multiple affiliations for an author, and at
least to allow the same to choose which publications pertain to which organization. In this way, both organizations and authors would have the right numbers. At now, as far as I understood, the latest organization takes all the publications of a certain author
if the author updates it, otherwise the author is put with an old affiliation, often missing a lot of publications.
Thank you very much for your attention,
P.S.: As noticed elsewhere, there is some big issue with citations since your numbers are very much different with respect to all the other engines, but honestly I cannot grasp the cause.