Incomplete parsing of bibliographic items by google scholar

I’m administrating the ToSC journal running OJS and our articles are successfully included in google scholar.

Recently I was told that some citations from articles published in our journal are not recognized by google scholar (while others are correctly listed in the cited paper’s citations).

On example I’m aware of is the paper The SKINNY Familiy of block ciphers and its low-latency variant MANTIS.
It is e.g. cited (and listed as such) by this article (which was published in ToSC) but also by this and at least three other articles, which where also published in ToSC, but which are not listed in the citations (and thus counted as) of the SKINNY block cipher.

Is anyone aware of other such examples? I assume that this is not a problem of OJS but of the published PDFs as GS needs to parse these PDFs to extract the correct bibliographic informations. Also if anyone has any ideas how to debug this further or what could help, it is greatly appreciated.

The google scholar and dublin core plugins are installed.

Hi @asante

Recommendations from Google Scholar:

there’s a bibliography section titled, e.g., “References” or “Bibliography” at the end.

Basically, you can add references in Submission and Publication Metadata Form for each submission. After parsing they will be available on the article landing page. Usually, this is enough (also it may take some time for Google bots to find them).

Other approach requires programming and implies adding citations embraced into Highwire Press tags inside <head> element on the article landing page, like, <meta name="citation_reference" content="...">.

I understood this in the following way: the articles PDF should contain such a section (which is the case for our submissions) - or is it referring to the articles webpage?

Thanks for the hint to the Submission Metadata Form, we’ll try this first.

I support your idea that capabilities of Google to parse PDF’s is limited and I don’t think that it will work in all cases. For example, it can depend on the method of PDF generation by the publisher - probably, PDF from LaTeX has more chances to be recognized as a text.