Tuesday, 27 November 2012

Google Scholar Annoyances

Last week, I posted about Google Scholar, and how it ranks areas of the arXiv above most journals in terms of a whole-journal h index.  More usefully, it keeps tracks of citations to my papers, so i can see whether what I'm doing is having any impact, and it suggests new articles to me that I might like to read.  It's suggested some that I'm sure I wouldn't have otherwise come across, from parochial journals, for example.

I try not to get too obsessed with the headline number, telling me my total number of citations.  After all, I can easily find someone I went to University with, whose single most cited paper has more citations than my entire lifetime's work.  And not just a bit more - a factor of 20 more.  Oh well.  One of the strange things, though, is that the number of total citations Google Scholar gives can go down as well as up over time.  The whole shebang is automated, of course, and a computer program finds all the citations and updates the database.  Presumably it sometimes roots out erroneous citations from the database.  The weird thing for me is that one recent paper of mine keeps jumping around in terms of number of citations.  It got up to around 11, then dropped to 0, then up to 5 or so, and today seems to have reverted to 0.  Even the Phys Rev C page for the article is aware of a few references, and it does not have a terribly extensive search remit, as far as I can tell.  Strange, and a little bit frustrating.  It'd be nice if it worked better than it does.