Discrete Mathematics and Big Data

(Re-posted from the BCC website)

On 15-17 February, we are holding a meeting at St Andrews on Discrete Mathematics & Big Data. You can find the web page here, and from that page you can get a programme for the event.

I had little to do with the organisation, though I was responsible for suggesting a couple of the speakers.

As I have said before, discrete mathematics can potentially generate huge amounts of data (though these tend to differ from most scientific data in that they are exact rather than approximate). Producing such data is clearly important and difficult, but there are further problems (storing it, curating it so as to make it useable by others, and so on) which have perhaps not been as much thought about as the production of the data. In the past, the usual thing was simply to put up a webpage with a link to the data.

I regard the ATLAS of finite group representations as a model of how this should be done. The data (generators of the groups in various permutation and matrix representations, character tables, etc.) is clearly laid out for human use, but (more importantly) is accessible by computer algebra programs such as GAP in a way which is practially transparent to users.

So I am very glad that Rob Wilson, the driving force behind the ATLAS, is speaking at the meeting (though he is not talking exclusively or even mainly about this – there is plenty more he has achieved in this area!)

Another speaker I am glad to welcome is Patric Østergård, one of the heroes of combinatorial search; among the big datasets he has been involved with producing is the catalogue of Steiner triple systems of order 19: there are 11084874829 of these up to isomorphism!

Come along if you can, and please contribute to the discussion, and help develop good practice for dealing with large combinatorial datasets.


About Peter Cameron

I count all the things that need to be counted.
This entry was posted in events and tagged , , , . Bookmark the permalink.

3 Responses to Discrete Mathematics and Big Data

  1. Dima says:

    A bit of shameless advertising: nowadays for each known to be feasible set of parameters of strongly regular graphs in Andries Brouwer’s webpage http://www.win.tue.nl/~aeb/graphs/srg/srgtab.html for at most 1300 vertices, Sage(math) will produce you an example of such a graph (with a dozen exceptions, some of them are actually seem to indicate wrong constructions in the literature).

    sage: g=graphs.strongly_regular_graph(231, 30, 9, 3); g

    Cameron Graph: Graph on 231 vertices

    cf. http://arxiv.org/abs/1601.00181
    Graphs from infinite series are as a rule provided as an implementation of the construction, rather than just as data.

    • Thanks Dima — very timely, today I start my Advanced Combinatorics lectures, and this year the first topic is strongly regular graphs! So I put a plug for your paper in the notes.

      • Dima says:

        We will create a Sagemath demo (actually, a “notebook” one can import and start using/changing/computing things) on cloud.sagemath.com to demonstrate few possibilities.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s