Category Archives: References And Citations

While implementing some improvements to our citation algorithm we introduced a bug which led to overcounting citations for about 2% of our records. We are in the process of fixing this now and are reindexing the citation data. This has caused some citation counts to drop initially. Numbers are going slowly up again and citation counts should be correct before the weekend.

Thanks for your patience and understanding.

As a further step towards the SPIRES shut off, citation curation is now done completely on INSPIRE. The INSPIRE reference extractor is more powerful than the SPIRES algorithm and hence, there will be more correct references extracted automatically already from day one. Furthermore, this means that we are able to show new citations faster now.

During the last weeks, SPIRES overwrote references extracted on INSPIRE which has led – as some of you noticed – to citation fluctuations. But all correct citations will be back again. Moreover, we will continue to manually add references that our extractor did not catch. And by the way – as a user of INSPIRE, you can help us by adding and correcting references yourself via the reference update form (https://old.inspirehep.net/help/reference_corrections).

Please send questions and feedback on references or other INSPIRE services and tools to feedback@inspirehep.net.

As some of you noticed, last week INSPIRE citation counts were momentarily unstable. This was due to the fact that, as part of the migration of data between SPIRES and INSPIRE, we cleaned up our citation database and deleted some “ghost citations”. This means either of two things had meanwhile happened: in a small number of cases the citing records were removed from INSPIRE or, most often, the citation list from arXiv pre-prints was updated to reflect either the published version of the article or the final re-submission of the pre-print. This is a common process which, with the migration between SPIRES and INSPIRE, had recently taken some backlog. Therefore, numbers might be slightly smaller now but no correct citation will be lost. In addition, better author disambiguation and authors claiming their papers help us calculate citation counts more accurately on INSPIRE.

Following the requests for more citation metrics we received we will soon provide more features for the citesummary.

For any questions and feedback on citations and other INSPIRE services, please contact us at feedback@inspirehep.net

Citations are a very core function of what we do; our users scrutinize them very closely and take the time to send us corrections. Thanks to this feedback we learned that some references and citation counts do not display correctly and we have been able to find the related bugs. It is our highest priority to solve these and we are working hard to fix them. The reasons for errors concerning references and citations are multifaceted:

• Citation links displayed with a paper don’t give reliable results or references that are visible on a paper’s reference list don’t show up as citations for the cited paper. This is a problem of the indexing function of INSPIRE, not the data. Fixing this is a top priority for us and we are currently working on it.
• The extracted reference list on a temporary entry misses a lot of references. We are working on a more robust reference extractor to address this issue, and we will reintroduce the %%CITATION in INSPIRE. This will help us extract references more accurately during this transition period, and as we expand our coverage of citations to books, conference papers and unpublished preprints.
• Citations to conference papers are disappearing. We have already addressed this issue, and are working on new reference extraction methods that will allow us to track citations to conference papers more robustly; we will do this retroactively of course.
• There has been a citation discrepancy between SPIRES and INSPIRE in many journals with volume letters. Much of this was cleaned up in SPIRES over the past six months already, but some discrepancies remain that are being given more focused attention right now.
• Incorrectly formatted cites in SPIRES with ambiguous journal codens do not get counted in SPIRES, but may be counted in INSPIRE if they have a valid volume and page. There are some discrepancies and we are working on it.
• We have a number of records that are blocked from updating in INSPIRE due to special processing of large author lists, or are otherwise curated currently only in Inspire. This will create a discrepancy in cite counts, for some records INSPIRE will be correct, and SPIRES incorrect. This is being addressed.

But don’t worry – no citation will be “lost” as we will reprocess the existing content. Once the bug fixing is done, we will also start introducing additional frequently requested data (e.g. the removal of self-citations). Please let us know if you have of any questions, feedback@inspirehep.net

And by the way – as a user of INSPIRE, you can also help providing correct publication and citation lists: First – we have come up with a feature that allows you to claim your papers so that your publication lists show correctly on INSPIRE (you just search for your name at http://inspirehep.net/person/search , choose your profile and follow the “verify your publications link”). Second – there is a new tool allowing you to correct references yourself (http://old.inspirehep.net/help/reference_corrections). Please check theses tools out and let us know what you think!

Citations are an important part of the service INSPIRE provides, as we know from your feedback. Ensuring that your citations are correctly counted requires an enormous amount of technical infrastructure and human effort in order to keep the reference lists of records current.   Not only do we parse references from many different formats in LaTeX and PDF, but papers are updated both at arXiv and in the process of publication in a journal, and of course there are plain old mistakes and omissions.

Over time at SPIRES we developed a mechanism for feedback if you find an error in our reference lists, but those of you who used it regularly know that it was very tedious to get all the information in the right format for us to enter easily.  Fortunately, INSPIRE brings us into the 21st Century with a new web-based form that allows you to add references to a record without cutting and pasting or having to know the abbreviations of journal names by heart.

For details on how to use this new form see:

http://inspirebeta.net/help/reference_corrections

This form should make it much easier to add missing references than our previous methods, and allows our INSPIRE team to approve your corrections quickly.  

Heath O’Connell

Fermilab

For as long as the annual topcited papers lists have been around, the all-time champion has been Weinberg’s “A model of leptons”, the 1967 paper that laid the foundation stone for the Standard Model. 30 years later, in November of 1997, the paper The Large N limit of superconformal field theories and supergravity by Maldacena appeared that established a connection between string theory and quantum field theory. It immediately set of a revolution in HEP and was the most highly cited paper ever since. Remarkably, its highest citation count was in 2010, where it received over 1,000 citations in a single year! One reason for this is the heavy ion results from Brookhaven that drew people to conclude that, based on Maldacena’s work, the quark gluon plasma can be modeled using string theory techniques.