Hi, there were several articles by Peter Jasco in the past regarding search quality in Google Scholar (GS), see for example
http://www.libraryjournal.com/article/CA6698580.html The GS "inclusion guidelines" are about three years old now, so I'm wondering about the discussion now. In the past, the number of documents from repositories was even higher in Google than in GS! The experience with our own repositories shows, that providing GS metadata clearly increased the number of documents of our repositories in GS, but is still below 50%. BASE covers repository content much better than GS, of course GS has other qualities (citation counts, library links, ...). >From my point of view the exciting question is, if GS uses the GS metadata only to get the fulltext easier from a repository or if GS uses the metadata in addition to the fulltext in order to improve the search quality within GS. The other question is, how much the Google/GS ratio of documents from repositories has changed in the last years. Best Dirk ------------------------------------------ Dirk Pieper Bielefeld UL - BASE Universitätsstr. 25, D-33615 Bielefeld E-mail: [email protected] | Tel.: +49 521 106-4010 Fax: +49 521 106-4052 www.ub.uni-bielefeld.de www.base-search.net ------------------------------------------ +++ Welcome to the 10th International Bielefeld Conference, 24. - 26. April 2012, http://conference.ub.uni-bielefeld.de +++ ----- Ursprüngliche Nachricht ----- Von: Stevan Harnad <[email protected]> Datum: Samstag, 18. Februar 2012, 8:31 Betreff: [GOAL] {Disarmed} Re: Google Scholar discoverability of repository content An: "Global Open Access List (Successor of AmSci)" <[email protected]> Cc: SPARC IR <[email protected]> > Begin forwarded message: > > From: Betsy Coles <[email protected]> > Date: February 17, 2012 5:48:42 PM EST > To: [email protected] > Subject: Re: [EP-tech] Re: Google Scholar discoverability of repository content > > I'm the technical manager for the main IR at Caltech, CaltechAUTHORS > (MailScanner has detected a possible fraud attempt from "authors.library.caltech.edu" claiming to be http://authors.library.caltech..edu), currently running EPrints 3.1.3.  > > Tim's conjecture 1) below seems to account almost exactly for the result > the article authors found: 87.7% of the 25,072 eprints in CaltechAUTHORS > have OA documents attached; the remainder have only documents that are > either restricted to campus or to repository staff.  I don't think there are very > many cases of Tim's conjecture 2), since we have concentrated on adding > current content. > > I haven't read the article in question (we don't subscribe), but the percentage > of open access eprints is almost exactly the same as the authors' report of GS > indexed items in Table 2.  I haven't tested specifically, but it's tempting to > conclude that GS is indexing 100% of our open access content. > > Betsy Coles > Caltech Library IT Group > [email protected] > > -----Original Message----- > From: [email protected] [mailto:[email protected]] On Behalf Of Tim Brody > Sent: Friday, February 17, 2012 3:33 AM > To: [email protected] > Cc: [email protected] > Subject: [EP-tech] Re: Google Scholar discoverability of repository content > > Hi All, > > Here is some specific advice for existing repository administrators from Google Scholar: > http://roar.eprints.org/help/google_scholar.html > > As far as I'm aware there isn't anyone running EPrints 2 now, so EPrints-based repositories are already (and for a long) the "best in class" for Google Scholar. > > > Right, this paper ... > > Table 1 is irrelevant and misleading. Scholar links first to the publisher and, only if there is no publisher link, directly to the IR version. That's a policy decision on the part of Scholar and nothing to do with IRs. > > Table 2 gives us some useful data. The headline rate for EPrints is 88% (based on CalTech). Unfortunately the authors haven't provided an analysis of what happened to the missing records. I've done a quick random sample of CalTech and I suspect the missing records will consist > of: > 1) Non-OA/non-full-text records (I'm sure a query to the CalTech repository admin could supply the data). > 2) A percentage of PDFs that Scholar won't be able to parse. CalTech contains some old (1950s), scanned PDFs from Journals. Where the article isn't at the top of the page Scholar will struggle to parse the title/authors/abstract and therefore won't be able to match it to their records e.g. http://authors.library.caltech.edu/5815/ > > > The remainder of the paper describes the authors' process of fixing their own IR (based on CONTENTdm). > > > The authors then wrongly conclude: > > "Despite GSâs endorsement of three software packages, the surveys conducted for this paper demonstrates that software is not a deciding factor for indexing ratio in GS. Each of the three recommended software packages showed good indexing ratios for some repositories and poor ratios for others." > > The authors looked at one instance of EPrints and, despite being a relatively old version, found 88% of its records indexed in GS. > > It is unfortunate that this paper has suggested that IR software in general is poorly indexed in GS. On the contrary, some badly implemented IR software is poorly indexed in GS. > > > After all that is said, the most critical factor to IR visibility is having (BOAI definition) open access content. Hiding content behind search forms, click-throughs and other things that emphasise the IR at the expense of the content will hurt your visibility. > > Lastly, Google will index your metadata-only records while Google Scholar is looking for full-texts. Your GS/Google ratio will approximate how many of your records have an attached open access PDF (.doc etc). > > > Sincerely, > Tim Brody > (EPrints Developer) > > On Wed, 2012-02-15 at 11:31 +0000, Stevan Harnad wrote: > Can we enhance the google-scholar discoverability of EPrints (and > DSpace) repositories? > > http://linksource.ebsco.com/linking.aspx?sid=google&auinit=K&aulast=Ar > litsch&atitle=Invisible+Institutional+Repositories:+Addressing+the+Low > +Indexing+Ratios+of+IRs+in+Google+Scholar&title=Library+Hi+Tech&volume > =30&issue=1&date=2012&spage=4&issn=0737-8831 > > Kenning Arlitsch, Patrick Shawn OBrien, (2012) "Invisible > Institutional > Repositories: Addressing the Low Indexing Ratios of IRs in Google > Scholar", Library Hi Tech, Vol. 30 Iss: 1 > > Purpose - Google Scholar has difficulty indexing the contents of > institutional repositories, and the authors hypothesize the reason is > that most repositories use Dublin Core, which cannot express > bibliographic citation information adequately for academic papers. > Google Scholar makes specific recommendations for repositories, > including the use of publishing industry metadata schemas over Dublin > Core. This paper tests a theory that transforming metadata schemas in > institutional repositories will lead to increased indexing by Google > Scholar. > > Design/methodology/approach - The authors conducted two surveys of > institutional and disciplinary repositories across the United States, > using different methodologies. They also conducted three pilot > projects that transformed the metadata of a subset of papers from > USpace, the University of Utah's institutional repository, and > examined the results of Google Scholar's explicit harvests. > > Findings - Repositories that use GS recommended metadata schemas and > express them in HTML meta tags experienced significantly higher > indexing ratios. The ease with which search engine crawlers can > navigate a repository also seems to affect indexing ratio. The second > and third metadata transformation pilot projects at Utah were > successful, ultimately achieving an indexing ratio of greater than 90%. > Research limitations/implications - The second survey was limited to > forty titles from each of seven repositories, for a total of 280 titles. > A larger survey that covers more repositories may be useful. > > Practical implications - Institutional repositories are achieving > significant mass, and the rate of author citations from those > repositories may affect university rankings. Lack of visibility in > Google Scholar, however, will limit the ability of IRs to play a more > significant role in those citation rates. > Originality/value - Little or no research has been published about > improving the indexing ratio of institutional repositories in Google > Scholar. The authors believe that they are the first to address the > possibility of transforming IR metadata to improve indexing ratios in > Google Scholar. > *** Options: > http://mailman.ecs.soton.ac.uk/mailman/listinfo/eprints-tech > *** Archive: http://www.eprints.org/tech.php/ > *** EPrints community wiki: http://wiki.eprints.org/ > > > *** Options: http://mailman.ecs.soton.ac.uk/mailman/listinfo/eprints-tech > *** Archive: http://www.eprints.org/tech.php/ > *** EPrints community wiki: http://wiki.eprints.org/ > > _______________________________________________ > GOAL mailing list > [email protected] > http://mailman.ecs.soton.ac.uk/mailman/listinfo/goal [ Part 2: "Attached Text" ] _______________________________________________ GOAL mailing list [email protected] http://mailman.ecs.soton.ac.uk/mailman/listinfo/goal
