3
$\begingroup$

I am a physicist working in TCS in a new group. I would like to know, given a conference (imagine your favorite ones), how can I know the quality of this conference and also the quality of the journal where the proceedings will be pusblished (in case they are). I heard something about CORE classification but found no link or more information about. Thanks

P.S.: I find TCS stack exchange very interesting for TCS discussions (obviously). I just wonder if there is any similar stack exchange academic site for physics/chemistry. I found physics.stackexchange, but some questions related with journals were instantly classified as offtopic

$\endgroup$
  • 1
    $\begingroup$ How do you (personally) judge the quality of physics journals and conferences? CS has a much larger emphasis on conferences than in physics, but the basic idea is the same. $\endgroup$ – Joe Fitzsimons May 30 '11 at 14:23
  • 1
    $\begingroup$ In physics, conferences are not so important. For journals we just have a look at the journal impact factor, usually or to the number of citations $\endgroup$ – Open the way May 30 '11 at 14:29
  • 3
    $\begingroup$ @gpu_drug: There is a physics stack exchange, but it is not research level. There is, however a proposal for a theoretical physics research level site here: area51.stackexchange.com/proposals/23848/theoretical-physics If you're interested, please do commit so that it goes into beta as sooner. $\endgroup$ – Joe Fitzsimons May 30 '11 at 14:47
  • 2
    $\begingroup$ Could you please remove the links to the specific examples of conferences? We would like to avoid doing any conference-ranking here. (Also note that the examples that you gave aren't really TCS conferences; some of them seem to be very applied – this web site focuses on theoretical computer science.) $\endgroup$ – Jukka Suomela May 30 '11 at 15:40
  • 6
    $\begingroup$ An easy rule of thumb: where are the good papers on your topic published (the ones you cite, or at least used in the preparation of your work)? $\endgroup$ – Sylvain May 30 '11 at 15:50
11
$\begingroup$

Usually, you should be able to figure out who are the leading researchers or research groups in your research area.

Everything else is then usually fairly straightforward: just find out in which conferences the leading researchers publish their work, and in which conferences they serve in programme committees, etc. Most likely those would be the most relevant conferences for your work as well.

The same applies to journals as well.

(Note that the most relevant conference is not necessarily the same as the most prestigious conference, but it might be a good idea to start with relevant conferences... Your work might have much more impact that way, even if it does not look that impressive in your CV.)

$\endgroup$
  • 3
    $\begingroup$ I agree Jukka. Ask a number of experts working in that field. See where the famous defining and groundbreaking papers of the field have been published. As an example, lets consider complexity theory, take a good book like Arora and Barak and check the bibliography, looking at a few pages you will see that STOC and FOCS stand out. This should give an idea about the top venues to publish. $\endgroup$ – Kaveh May 30 '11 at 23:27
10
$\begingroup$

See my previous answer about journals. Same thing.

Different communities withinin theory (like in every other subject) have vastly different opinions about the relative quality of publication venues. The only way to judge the quality of a conference is to judge the quality of the research that is presented and published there, as viewed by the sub-community you want to impress.

If the only community you're trying to impress is a hiring/promotion committee, you're doing it wrong. To get hired/promoted, you need strong letters from experts in your field that are recognized as such by hiring/promotion committees. Figure out who those people are, and then go kick ass on their home turf.

$\endgroup$
9
$\begingroup$

One possible ranking scheme is the so-called Australian Ranking of Computer Science Conferences: At the bottom of this page you'll find various lists containing the data organised in different ways. For example, this one is ordered by acronym, and is probably the most useful. If the conference is not in the list, it is probably new and thus neither established or necessarily high quality.

Regarding the quality of the publication, the SCI index can tell you that sort of information. But this is unreliable for conferences published in LNCS, as some volumes are included and some are not. (There are tools and websites providing SCI data.)

Other things to look for are the number of editions the conference has had, whether it is a conference, symposium or workshop, whether it has ACM, IEEE, or other such sponsorship. None of these are particularly reliable metrics, though.

Finally, ask your colleagues in the field.

$\endgroup$
  • 2
    $\begingroup$ For this one, keep in mind that it has a strong bias since it has been done by researchers from the same geographic area. Thus, some research fields are under represented due to this bias. $\endgroup$ – Sylvain Peyronnet May 30 '11 at 17:33
2
$\begingroup$

Although not strictly a measure of quality, one easy way to judge the perceived quality of a conference is to look at its historical acceptance rate. In Computer Science, many conference publications are considered terminal because they are so competitive. The idea is that the "better" the conference, the more desirable it is to submit one's work, thus the more submissions it receives, thus lowering its acceptance rate. It is common for some highly competitive conferences to have acceptance rates well below 20%. If we believe in the efficacy of peer review (which may be a big "if"), then the lower acceptance rate should correlate with "better" papers.

$\endgroup$
  • 1
    $\begingroup$ This is certainly one of the measures the bean-counters use when assessing their academics. $\endgroup$ – Dave Clarke Jun 1 '11 at 12:16
-2
$\begingroup$

I recommend Microsoft Academic Search. See for example: a ranking list for TCS conferences

$\endgroup$
  • 2
    $\begingroup$ I've tried it before (checking my own citations), and they were way way off, so I'm not sure how much I would trust their ranking at the moment. $\endgroup$ – Joe Fitzsimons May 30 '11 at 16:38
  • $\begingroup$ This ranking may not be good for physicists, but I, as a computer scientist, think that rankings for computer science are fairly reliable. $\endgroup$ – Snowie May 30 '11 at 16:50
  • 3
    $\begingroup$ @Snowie: I just quickly checked the top-ranking conferences in my research area, distributed & parallel computing. Number 1 conference seems to be something called "Java Grande"... Let's not try to rely on any automatically-generated conference ranking. $\endgroup$ – Jukka Suomela May 30 '11 at 17:00
  • 2
    $\begingroup$ @Snowie: It's just as bad for my TCS publications as my physics ones. $\endgroup$ – Joe Fitzsimons May 30 '11 at 17:13
  • 2
    $\begingroup$ Some of the things listed as conferences are book series. And the prestige ranking, while not completely wrong, isn't very good, either. $\endgroup$ – Peter Shor May 30 '11 at 19:49

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.