Dear all,
we would like to use a one-sentence-one-record (OROS) - version of the tagged
COLT corpus for our current research. As with all corpora of this size the
conversion process requires careful planning as well as reliable test routines
to make sure nothing gets lost during the conversion process. Has anyone
successfully converted the original tagged version of COLT (ICAME CD, Second
Edition, 1999) to an OROS version and could give us some hints on the procedure
and/or would be willing to share some reliable (awk / perl) scripts?
Regards,
Norbert Schlüter
-----------------------
Freie Universität Berlin
nosch@zedat.fu-berlin.de
This archive was generated by hypermail 2b29 : Thu Jun 12 2003 - 16:04:23 MET DST