Textpatterns in a computer assisted translator's workstation

Abstract

A software package for a computer-assisted translator's workstation should contain a special module which consists of a database of preferred textual structures in the source and target languages, (TEXTPAT I), as well as a processor of typical translation cases (TEXTPAT II). TEXTPAT I includes micro- and macrostructures at four levels (text type, text type variants, chunks, syntactic and lexical structures). TEXTPAT II consists of lists of items for which translation rules have to be applied. Both textpats contribute to the idea of a translator's expert system.