Difference between revisions of "Apertium-kaz-tat/paper"

From Apertium
Jump to navigation Jump to search
Line 2: Line 2:
   
 
== TODO ==
 
== TODO ==
  +
Ideal benchmarks:
  +
* document rules in the rlx with example sentences
  +
* more like 100-150 (currently ~40) disambiguation rules in -kaz
  +
 
=== Ilnar ===
 
=== Ilnar ===
 
* Work on increasing <s>coverage (via lexc) and</s> trimmed coverage (via dix) to 90%
 
* Work on increasing <s>coverage (via lexc) and</s> trimmed coverage (via dix) to 90%

Revision as of 19:51, 1 April 2013

We're submitting a paper on apertium-kaz-tat to MT Summit 2013. DEADLINE: APRIL 20.

TODO

Ideal benchmarks:

  • document rules in the rlx with example sentences
  • more like 100-150 (currently ~40) disambiguation rules in -kaz

Ilnar

  • Work on increasing coverage (via lexc) and trimmed coverage (via dix) to 90%
  • Work on making sure testvoc passes
  • Later: correct some tests for error-rate testing

Fran

  • Delegate out error-rate testing tasks

JNW

  • Work on last few issues in -tat twol
  • Write up background, etc. for paper