Pete Whitelock

Learn More
In this paper, I describe a view of Machine Translation (MT) that promises radical new solutions to some of the problems of structural approaches such as transfer and interlingua. The new approach is based on a lexicalist view of g rammar in which lexical i tems are mult idimensional signs. Translation equivalence is defined between bags (multi-sets) of(More)
At the Centre for Computational Linguistics, we are designing a n d implementing an English-to-Japanese interactive machine translation system. Tile project is funded jointly by t h e Alvey Directorate and International Computers Limited (ICL). 'Pile prototype system runs on the ICL PERQ, though much of the development work has been done on a VAX ]]./750.(More)
This paper discusses the application of Unification Categorial Grammar (UCG) to the framework of Isomorphic Grammars for Machine Translation pioneered by Landsbergen. The Isomorphic Grammars approach to MT involves developing the grammars of the Source and Target languages in parallel, in order to ensure that SL and TL expressions which stand in the(More)
We present the design of a practical context-sensitive glosser, incorporating current techniques for lightweight linguistic analysis based on large-scale lexical resources. We outline a general model for ranking the possible translations of the words and expressions that make up a text. This information can be used by a simple resource-bounded algorithm, of(More)
This paper describes the Sharp Intelligent Dictionary (SID), an English-Japanese glossing system for Japanese readers and learners of English. SID uses a variety of lightweight analysis techniques, a large bilingual dictionary and a prioritised model of collocations to present informed guesses about the best translations of words and expressions in their(More)
The translation quality of current machine translation systems is limited by restriction to syntactic and semantic factors. A wider range of factors, including pragmatic and discourse factors, needs to be extracted from the source text and transferred to the target text. However, correct generation of the target text often requires factors which are not(More)
We present the design of a practical context-sensitive glosser, incorporating current techniques for lightweight linguistic analysis based on large-scale lexical resources. We outline a general model for ranking the possible translations of the words and expressions that make up a text. This information can be used by a simple resource-bounded algorithm, of(More)