The following examples are a sample of the commercial applications we have or are currently helping to build:
  • We helped Corpora (now Infonic) build their sentiment classifier.
  • We helped Texperts (now kgb) develop techniques for efficient processing of SMS Text Questions.
  • We are sharing technology with TouchType to develop the next generation of text prediction software for smartphones.
  • We are participating in the English Profile Project helping Cambridge Assessment and Cambridge University Press leverage learner corpora to develop the next generation of EFL teaching and testing materials.
  • We have developed an automated ESOL text grading system and are working with Cambridge ESOL to deploy it.
  • We are developing a scientific text mining system by converting PDF papers to SciXML, parsing them with RASP, and then indexing them with Solr/Lucene, building on the FlySlip project in collaboration with Camtology.
  • We have developed an anonymisation tool capable of supporting more rapid and accurate anonymisation of sensitive data such as patient medical records (Demo) and are working with GPRD (a part of the National Health Service) to deploy it.
  • We are working with Cisco to improve terminology recognition in Cisco Pulse.