Language engineering
Keywords Cybernetics, Language
A recent meeting at Baslow, Derbyshire UK, saw 20 participants from seven countries explore some of the problems of language engineering. This was a Workshop meeting that appreciated that the ability to extract meaning and select relevant material from information sources will become increasingly important in the digital age. It was realised that although language engineering as a field will provide the tools to achieve this, incompatibilities in such a diverse and fast moving area of endeavour can make it difficult to use these tools efficiently and effectively. This meeting explored a promising route to solving this problem: common architecture.
When the Workshop delegates discussed this they considered in particular:
The role of architectures in the examination, dissemination, use and evaluation of linguistic resources.
Use and evaluation of linguistic resources.
The relationship between general natural language processing (NLP) and those created for specific activities, like text generation.
The relationship between generic architectures and text mark-up languages.
Some of the problems highlighted included the way NLP researchers are generating tools for specific tasks, which limits take-up by other groups. This,it was believed, could be overcome by having more general tools, as in other fields. Lesser-used languages also produced particular problems.
The Workshop was organised by Professor Yorick Wilks (Sheffield University,UK). In summary he said that:
It was interesting to see the pragmatic approach being adopted, with groups dealing with the diversity rather than conforming to a particular standard.
This is a field that is particularly relevant to cyberneticians, who may find more details of the Workshop and related subjects on the Web: www.dcs.shef.ac.uk/~hamish/dalr/
