A new £600,000 project which could benefit users of the Amazon Kindle and other e-book reading devices is being developed by experts at the University of Wolverhampton.
Researchers aim to develop technology that could enhance the reading experience by clarifying the meaning of ambiguous words according to the different contexts in which they are used.
The study, which has been awarded £609,718 by the Arts and Humanities Research Council, will also benefit language teachers, linguists, and developers of computer software for language processing such as the semantic web.
Led by Professor Ruslan Mitkov, Director of the University’s Research Institute of Information and Language Processing (RIILP), the ‘Disambiguation of verbs by collocation (DVC)’ project will begin in October.
Professor Mitkov said: “E-book reading devices such as Amazon Kindle enable a dictionary to interact with any text that a user is reading. For such an application to be fully effective it is also necessary to enable the software to select not only the most relevant word, but also the most relevant sense of that word.
“Currently available research on word-sense relevance is not satisfactory. By embedding technology like the one developed in this project into e-book readers, it will be possible to enhance the reading experience.
“The project will also benefit language teachers, particularly those working in distance learning, linguists working on linguistic theories, lexicographers and people working in the field of computational linguistics.”
A pilot project by Professor Patrick Hanks from the University analysed 700 verbs and found that while words may be highly ambiguous, patterns of words are rarely ambiguous and most verbs can be assigned unambiguously to a pattern.
Using computational linguistics techniques, which focus on the processing of human languages by computers, the DVC project will analyse 3,000 common English verbs, and carry out in-depth analysis of 100 verbs.
The three-year project will show the comparative frequency of each pattern of each verb, enabling computer programs to develop statistically-based reasoning about meanings, rather than trying to evaluate all possibilities equally.
A new website will be developed to host interactive demos that will enable visitors to see the patterns collected and test the technologies developed.
It will also provide news about the progress of the project and links to research papers.
For further information, visit the project website: http://clg.wlv.ac.uk/projects/DVC/
For more information please contact Vickie Warren in the Media Relations Office on 01902 322736