I am looking for a sepcific English corpus that
allows to identify whether a translation system
is a machine translation system (i.e. integrating
classic linguistic mechanisms) or if it is a
translation memory based one. Typically, if some
word groups, in different sentences, are always translated
in the same way, while they should not, in this case,
the system would probably be a translation memory one.
Ideally the corpus I am loking for could be a set of
sentences that are selected or built for this purpose.
Thanks in advance.
France Telecom R&D
This archive was generated by hypermail 2b29 : Mon Dec 10 2001 - 09:12:48 MET