Press release on our ICWSM 13 paper

Posted on 4 July 2013 by Miles     

Edinburgh University press release talking about the relationship between Newswire and Twitter

Newswire FSD Corpus

Posted on 17 Apr 2013 by Luke     

You can download the corpus of newswire documents used in our ICWSM 2013 paper "Can Twitter replace Newswire for breaking news?" from here.
If you use the corpus in your work please cite that paper
See below for the corpus of tweets used.

Can Twitter replace Newswire for breaking news?

Posted on 26 Mar 2013 by Richard     

Preprint

Sasa Petrovic, Miles Osborne, Richard McCreadie, Craig Macdonald, Iadh Ounis and Luke Shrimpton

In Proceedings of ICWSM'13. Boston, USA. July 8-11 2013.

Code to download FSD Corpus

Posted on 25 feb 2013 by Miles     

Emaad Ahmed Manzoor has released some code allowing people to download from Twitter our Twitter FSD corpus. His code can be found on Github.

Cpp wrapper for Storm

Posted on 8 aug 2012 by Sasa     

We are releasing our Storm C++ wrapper that enables writing bolts in C++ and using them in Storm. The code can be found on Github. We have a short document describing how to use this wrapper.

Bieber no more: First Story Detection using Twitter and Wikipedia

Posted on 1 aug 2012 by Richard     

Preprint

Miles Osborne, Sasa Petrovic, Richard McCreadie, Craig Macdonald and Iadh Ounis

In Proceedings of TAIA'12. Portland, Oregon, USA. August 2012.

FSD corpus of tweets

Posted on 12 jun 2012 by Sasa     

We are releasing the corpus of first stories used in our NAACL 2012 paper "Using paraphrases for improving first story detection in news and Twitter". If you use the data in your work, please cite that paper. The corpus also contains user screen names which should enable fast crawling.

Download the corpus here (note that the file is 700 MB).

Talk at DCU, Ireland may 25th 2012

Posted on 24 may 2012 by Miles     

On Friday I will be talking about our past and ongoing work on FSD in Twitter.

Here are the slides .

S4 v Storm

Posted on 7 may 2012 by Richard     

We looked at S4 and Storm and came to the conclusion that Storm would be better suited to our project.

Here are some notes comparing the two approaches

Using paraphrases for improving first story detection in news and Twitter

Posted on 25 apr 2012 by Miles     

Preprint

Sasa Petrovic, Miles Osborne and Victor Lavrenko

NAACL, Montreal, Canada. June 2012