James Clarke

Current

• Postdoctoral Researcher at University of Illinois at Urbana-Champaign

Past

• PhD Student at The University of Edinburgh

Education

• The University of Edinburgh
• The University of Sheffield
• University of Exeter

see less...

Recommended

1 person has recommended James

Connections

83 connections

Industry

Research

James Clarke’s Summary

James Clarke’s Specialties:

Natural Language Processing, Computational Linguistics, Automatic Summarisation, Machine Learning

James Clarke’s Experience

• Postdoctoral Researcher

  University of Illinois at Urbana-Champaign

  (Educational Institution; 10,001 or more employees; Higher Education industry)

  June 2008 — Present (1 year 6 months)

• PhD Student

  The University of Edinburgh

  (Educational Institution; 1001-5000 employees; Research industry)

  October 2004 — June 2008 (3 years 9 months)

  Developing novel frameworks for automatic summarisation and sentence compression. Covering many aspects of compression including annotation methods, data analysis, models, frameworks and evaluation techniques. Formalisation of the compression task within an Integer Linear Programming Framework in which models can be supplemented with linguistic constraints.

James Clarke’s Education

• The University of Edinburgh

  PhD , Informatics , 2004 — 2008

• The University of Sheffield

  MSc , Advanced Computer Science , 2003 — 2004

  Dissertation on clustering techniques for multi-document summarisation. Involved an investigation into the influence of different clustering techniques for theme detection used during multi-document summarisation. Explored: pattern representations, similarity measures and clustering algorithms.

• University of Exeter

  BSc , Computer Science , 2000 — 2003

  Graduated with a first class degree and honours. Final year project involved the research, design and implementation of a multicast file transfer program with rsync capabilities. This resulted in a program capable of synchronising files from a source machine to many clients at once efficiently using multicasting. Involved the creation of a reliable and scalable multicast protocol and modification of the rsync algorithm to work efficiently over multicast.

Additional Information

James Clarke’s Groups:

ACL (Association for Computational Linguistics)

James Clarke’s Honors:

Best Paper Award, EMNLP-CoNNL 2007

James Clarke’s Contact Settings

Interested In:

• career opportunities
• consulting offers
• new ventures
• expertise requests
• reference requests
• getting back in touch