The UCREL Corpus Research Seminar (CRS) is a forum for all staff, visiting academics, and postgraduate research students interested in corpus-based research in any area of linguistics. CRS is run by UCREL (University Centre for Computer Corpus Research on Language), a research centre between the Department of Linguistics and English Language and the School of Computing and Communications.
CRS meetings offer an opportunity to present work in progress and receive helpful feedback, discuss relevant research, approaches and methods, get experience in using corpus interfaces and tools, and stay up to date with corpus-based research at Lancaster University. We welcome anyone who is a newcomer to this exciting and growing area of linguistics. We welcome presentations from researchers of other departments and universities.
On this site, along with general information, you will find a list of upcoming seminars and an archive of past seminars. If you have any suggestions of things to add to the site, then please get in touch.
In 2014/2015 CRS meetings will be on Thursdays at 1pm during term time, unless otherwise indicated.
Notifications of seminars are sent to the UCREL Mailing List, sign up if you would like to receive them and other UCREL related messages. You can also follow us on Twitter, where we post updates on upcoming seminars.
If you need more information or want to give a presentation, please contact one of the CRS organisers:
[...] = add @lancs.ac.uk for email address.
This website is still under construction. The previous website has an archive of past seminars.
We acknowledge the following funding for external speakers: CASS and UCREL research centres, the Faculty of Arts and Social Sciences, the Department of Linguistics and English Language and the School of Computing and Communications.
The UCREL Corpus Research Seminars this academic year (2014/15) will be on Thursdays 1pm-2pm during term time, unless otherwise indicated, please check our upcoming page for the time and location of any future presenations. If you would like to give a talk or have a suggestion for an external speaker to invite, then please get in touch.
Thursday 30th October 2014
Furness LT 3
Social Media Analysis with GATE
Diana Maynard has been a Research Fellow at the University of Sheffield, UK since February 2000, after acquiring a PhD in Natural Language Processing from Manchester Metropolitan University. Her main interests are in information extraction, opinion mining, sarcasm detection, social media, terminology and semantic web technologies. She is the chief developer of Sheffield University's open-source Information Extraction tools, and currently leads the work on Information Extraction and Opinion Mining on the EU DecarboNet project. She is chair of the annual GATE training course, and leads the GATE consultancy on IE and opinion mining. She has published extensively, organised a number of national and international conferences, workshops and tutorials, taught at summer schools, and given many invited talks and keynotes.
Can Twitter predict my chances of getting flu? Is social media causing deterioration in English skills? How much should I trust hotel reviews online? What's the best helpline to use in Who Wants to be a Millionaire?
These days there is an increasing need to interpret and act upon information from large-volume, social media streams, such as Twitter, Facebook, and forum posts. While analysis of longer documents (e.g.
news) has been very well studied, understanding social media content has only recently been addressed. Social media poses three major computational challenges, dubbed by Gartner the 3Vs of big data: volume, velocity, and variety. Text analysis methods, in particular, face further difficulties arising from the short, noisy, and strongly contextualised nature of social media. To address the 3Vs of social media, novel language technologies have emerged, e.g. using locality sensitive hashing to detect new stories in media streams (volume), predicting stock market movements from tweet sentiment (velocity), and recommending blogs and news articles based on users' own comments (variety). In this talk I will discuss some of the main problems for social media language analysis, and explain how the GATE tools can be used to deal with them, based around the core topics of information extraction and sentiment analysis.
Week 6: 13th November 2014 (1:00-2:00pm)
Furness LT 3
The Aston Corpus of West Midlands English: a variationist approach to spoken data collection and analysis
Week 8: 26th November 2014 (4:00-5:00pm)
County South C89
Big Educational Data: any good for SLA research?
Week 10: 27th November 2014 (1:00-2:00pm)
Furness LT 3
A corpus-based bimodal analysis of referential behavior to inanimate objects