Reporter 401, 6 May 1997
Parsing sentences may have been a daunting task at school, but now it can all be done by computers. An experimental prototype developed at the University provides a free service which analyses text and splits it into its various grammatical parts.
Developed in the Leeds Centre for Computer Analysis of Language and Speech in the School of Computer Studies, the programme can receive text by email and then break down sentences to their component parts. There are several different theories of English grammar, and users of the service can choose which to apply.
A 'part-of-speech' tag is attached to each word, taking its context into account. In the sentence 'the quick brown fox jumped over the lazy dog', the programme will attach the tag common singular noun to 'fox' and adjective to 'lazy'.
Since January the service has had over 500 users, including publishers of English language teaching dictionaries and researchers investigating its uses in speech recognition. The tagging service was developed in an Engineering and Physical Sciences Research Council project and providing it over the internet allows a large audience to be reached. "Our service is different because it is available by email and offers a choice from eight standard part-of-speech tag categories," says Centre Director Eric Atwell.
[Main news stories | University home page | Events]