About me

I'm currently a post-doc at the University of Edinburgh (in case the hosting of this page didn't give location away!). I hold a BSc in Computer Science and AI from the University of Sheffield (graduated 2004), and a PhD in Natural Language Processing from the same place (in 2009). In my spare time I'm an avid cinema fan, I love music (live and recorded), anybody who knows me would say I'm obsessive about food (I love to cook), and I enjoy playing a whole range of sports. I'm originally from York, went to school in Tadcaster (a little town halfway between York and Leeds, famous for little more than the fact that it has four breweries which made for eventful entertainment and a strange smell) and moved to Sheffield at the tender age of eighteen where I got my undergrad and doctorate. Seven years later I moved to Edinburgh, where I still reside.

Research Interests

I got my Phd in (statistical) classification methods for documents, encompassing related methods in statistics, machine learning and language processing, from the University of Sheffield. I'm now working on problems at the intersection of computer vision and language, particularly detecting out of context regions of images and modelling the effect these have on human viewing behaviour. I have also worked on distributional models of word meaning, and detecting outliers (anomalies) in document collections, as well as developing and adapting language processing techniques for use with Arabic, Persian and Chinese. I am currently on the SynProc (synchronous processing) project at Edinburgh; I have previously been involved in several MoD funded projects, including the REVEAL and CLUE contracts at Sheffield.



  • Enron Authorship corpus: a 9-way authorship attribution problem derived from the Enron corpus (from the LREC 2008 paper above) Download(.tar.gz)
  • A corpus of Business and Personal mails for classification (from the ACL 2006 paper above). Download(.tar.gz)
  • For those who are really interested, take a look at my cv


Ben Allison
School of Informatics
Informatics Forum
10 Crichton Street

Tel.: +44 (0)131 651 5661
Email: (my first initial)(my surname) {at} inf.ed.ac.uk
(note there are no spaces or punctuation between the first intial and the surname)