Difference between revisions of "Talk:Earley-based structural transfer for Apertium"
Jump to navigation
Jump to search
(Created page with 'The current transfer assumes lemma/PoS/sub-pos (and possibly lexical selection) are all completely disambiguated. This reduces parse ambiguity _a lot_. It might even reduce it so…') |
|||
| Line 1: | Line 1: | ||
The current transfer assumes lemma/PoS/sub-pos (and possibly lexical selection) are all completely disambiguated. This reduces parse ambiguity _a lot_. It might even reduce it so much that it's not worth it to do parse selection, but simply choose the first, longest match, similar to what transfer currently does. This might make writing a recursive parse module a bit simpler since the Earley parser wouldn't have to keep track of shorter matches, and would not have to do anything after parsing to select between them. It's an empirical question how well that would work though (most likely it would work a lot better for pairs that use syntactic/dependency CG's). --[[User:Unhammer|unhammer]] 13:20, 2 October 2013 (UTC) |
The current transfer assumes lemma/PoS/sub-pos (and possibly lexical selection) are all completely disambiguated. This reduces parse ambiguity _a lot_. It might even reduce it so much that it's not worth it to do parse selection, but simply choose the first, longest match, similar to what transfer currently does. This might make writing a recursive parse module a bit simpler since the Earley parser wouldn't have to keep track of shorter matches, and would not have to do anything after parsing to select between them. It's an empirical question how well that would work though (most likely it would work a lot better for pairs that use syntactic/dependency CG's). --[[User:Unhammer|unhammer]] 13:20, 2 October 2013 (UTC) |
||
| + | |||
| + | ::Check out [[recursive transfer]] :) - [[User:Francis Tyers|Francis Tyers]] 21:20, 2 October 2013 (UTC) |
||
Latest revision as of 21:20, 2 October 2013
The current transfer assumes lemma/PoS/sub-pos (and possibly lexical selection) are all completely disambiguated. This reduces parse ambiguity _a lot_. It might even reduce it so much that it's not worth it to do parse selection, but simply choose the first, longest match, similar to what transfer currently does. This might make writing a recursive parse module a bit simpler since the Earley parser wouldn't have to keep track of shorter matches, and would not have to do anything after parsing to select between them. It's an empirical question how well that would work though (most likely it would work a lot better for pairs that use syntactic/dependency CG's). --unhammer 13:20, 2 October 2013 (UTC)
- Check out recursive transfer :) - Francis Tyers 21:20, 2 October 2013 (UTC)
