Machine Learning and Natural Language Processing Groups
Department of Computer Science
University of Sheffield
Telephone: +44 (0114) 2221825
Address: Room 148, Regent Court, Portobello Street, Sheffield S1 4DP, UK
My research interests are in natural language processing (NLP) and machine learning, focusing on graphical models for structured prediction and Bayesian inference.
Applications include automatic machine translation, parsing, language modelling, summarisation and information extraction.
Before September 2009, I worked at Edinburgh University as a Research Fellow for Mirella Lapata on models of text rewriting (summarisation, sentence compression and translation). Before that (2003 - 2006) I
did my PhD at Melbourne University supervised by Steven Bird and
Miles Osborne on developing novel
approximate inference algorithms for Conditional Random Fields (CRFs) to enable
their application to various NLP tasks.
We are hiring:
I received a best paper awards at CICLING 2013 and ACM Hypertext
2013, details below.
Please visit Reducing Annotation Effort for Quality Estimation via Active Learning
Daniel Beck, Lucia Specia and Trevor Cohn
ACL 2013 (short papers)
QuEst - A translation quality estimation framework
Lucia Specia, Kashif Shah, Jose Guilherme Camargo de Souza and Trevor
ACL 2013 (demonstrations)
Modelling Annotator Bias with Multi-task Gaussian Processes: An
Application to Machine Translation Quality Estimation
Trevor Cohn and Lucia Specia
A user-centric model of voting intention from Social Media
Vasileios Lampos, Daniel Preotiuc-Pietro and Trevor Cohn
An Infinite Hierarchical Bayesian Model of Phrasal Translation
Trevor Cohn and Gholamreza Haffari
Markov Translation using Non-parametric Bayesian Inference
Yang Feng and Trevor Cohn
Adaptation of lecture speech recognition system with machine translation output Raymond Ng, Trevor Cohn, and Thomas Hain ICASSP 2013
Mining User Behaviors: A study of check-in patterns in Location Based Social Networks Daniel Preotiuc-Pietro and Trevor Cohn. ACM Web Science 2013
Where's @wally: A classification approach to Geolocating users based
on their social ties Dominic Rout, Daniel Preotiuc-Pietro,
Kalina Bontcheva, and Trevor Cohn. ACM Hypertext 2013
Ted Nelson Award for Best New-comer
BLEU deconstructed: Designing a Better MT Evaluation Metric Xingyi Song, Trevor Cohn, and Lucia Specia. CICLING 2013
Topic-oriented Words as Features for Named Entity Recognition
Ziqi Zhang, Trevor Cohn, and Fabio Ciravegna. CICLING
2013 Best paper award winner (2nd place) this page for a full
list of my publications.
2011-2013 Barista: Non-parametric Bayesian Induction for Machine Translation, EPSRC First grant
2012-2013 SLATr: Spoken Language Translation, Google Research Award
2011-2014 TrendMiner: Large-scale, Cross-lingual Trend Mining and Summarisation of Real-time Media Streams, EU FP7 STREP
2012-2012 QuEst: an open source tool for machine translation Quality Estimation, PASCAL2 Harvest
2012-2012 Grammar Induction Challenge, PASCAL2 Challenge
2010 Johns Hopkins Summer Workshop: Unsupervised learning of translation grammars, NSF & Google
2007-2010 Discriminative Phrase-Based Statistical Machine Translation, EPSRC Resources
I ran a workshop on inducing structure from text at NAACL:HLT
2012 which included a PASCAL challenge competition on Grammar and
Part-of-Speech induction (see the webpage for the datasets.)
I have three post-docs, working machine translation and social network analysis projects.
I currently supervise the following PhD students:
I'm always on the lookout for new students. Please email me if you're interested.