Talk:How to get started with lexical selection rules

From Apertium
Jump to navigation Jump to search

It would be cool to have a program to calculate the most frequent errors in lexical selection from:

  • Parallel corpus
  • Tag both sides
  • Translate with defaults
  • Translate up to lexical transfer
  • See which words are most frequently mis-chosen.

- Francis Tyers 13:52, 7 December 2011 (UTC)