quinta-feira, 25 de novembro de 2010 10:46Hi,
I think it is very interesting that for each author you compute h-idex and g-index. However, it would be even more interesting (and pretty easy for you) if you also computed the h_m index described in
M. Schreiber, "To share the fame in a fair way, h_m modifies h for multi-authored manuscripts”, New Journal of Physics, 2010. (see http://iopscience.iop.org/1367-2630/10/4/040201/)
Given an author, h_m takes “multiple co-authorship appropriately into account, by counting each paper only fractionally according to (the inverse of) the number of authors.” This is IMHO an obviously and sorely needed normalization of the h index, which other tools for measuring scholarly output (e.g. Scholarometer) also compute.
Regards Fabrizio Sebastiani
Todas as Respostas
domingo, 28 de novembro de 2010 00:51Proprietário
thank you for your suggestion. we have thought about considering the number of co-authors when measuring the contribution of an author, we also understand that authors in different fields (different journals) have different pactice and tradition of number of co-author. we have to run more experiements and get people's feedback first.
in the long run, we want to open up our meta data so that different people can measure our entities (author, paper, organization) differently according to their preference.
- Marcado como Resposta Cherry CHEMicrosoft Employee, Owner terça-feira, 21 de dezembro de 2010 05:45