RSS feed
<< Service Upgrades | Home | Getting Document XML >>

CiteSeerX Beta

We have just released a major update to the CiteSeerX service.  The most significant changes are under the hood, including better citation matching and improvements to the data model; however, there are several new features and UI improvements that may prove useful.

Specific enhancements of note include the following:

  • BibTeX records now appear on document summary pages for convenient copy/paste access.
  • All document file formats are now downloadable.  This means that if a document started as PostScript and was later converted to PDF, both formats will be available.  However, if a document was originally found in PDF format, only the PDF will be available.
  • Self citations are now tracked.
  • All search queries can now be converted to ATOM and RSS feeds.  This feature is particularly handy when combined with the advanced search form.  For example, you can track new content by author, institution, or text phrase or track new citations to individual papers and get updates right in your email inbox.  Look for the ATOM and RSS icons on search result pages.
  • MyCiteSeer has gone through several upgrades.  For example, Bulletin announcements now show up on the account landing page and some new features have been added to the collections module.

Next on the agenda is enabling user corrections to document metadata and bringing the submission system online.  We are working to add these features as soon as possible.

As always, feedback is welcome.


Re: CiteSeerX Beta

Hi Isaac.

The RSS feature looks great.

I have one proposition regarding "self citations": For now there is only one link, e.g. "15 citations - 6 self". Would it be possible to break it into two links? "15 citations" as before and "6 self" showing only self citations.

One minor bug: to logo of this blog is still "CiteseerX alpha" and the page title of the main page, too.

Re: CiteSeerX Beta

Hi Isaac,

I don't know what has happened to the BibTeX data, but their quality is much worse than before the "X" upgrade. Often, fields are missing and the respective data is merged into other fields. For example (CiteseerX):
@MISC{Brown_,
author = {Duane C. Brown},
title = {},
year = {}
}
Old Citeseer (from Google cache, which is fortunately still working):
@article{ brown71closerange,
author = "D. C. Brown",
title = "Close-range camera calibration",
journal = "Photogrammetric Engineering",
volume = "37",
pages = "855--866",
year = "1971",
url = "citeseer.ist.psu.edu/583868.html"}

You should really do something about this!

Best regards,
Thomas Stehle

Re: CiteSeerX Beta

hi everybody who's working at citeseerX. i agree about the division of self citations from others citations. furthermore, i'd like to ask one thing i don't understand. if I search for "statecharts" i find a bunch of resulting papers. if i click on "on the semantics of statecharts" (the first one, for an example) i can see among the cited articles and books the entry "Statecharts: A visual formalism for complex systems" (the first one) and from there i can see how many articles cited "Statecharts: A visual formalism for complex systems". but what if i search directly "Statecharts: A visual formalism for complex systems"? i can't find any result! i tried including citations in the search, and also i tried the advanced search, without getting any result. is this a normal behavior?!? thanks for the attention

Re: CiteSeerX Beta

bibtex entries are not very useful if they only have the author's name. citeseer had much more complete entries. Why the regression? Can the citeseer implementation be added to citeseerx?

Add a comment Send a TrackBack