[ Skip to the content ]

Institute of Formal and Applied Linguistics Wiki


[ Back to the navigation ]

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revision Previous revision
Next revision
Previous revision
Next revision Both sides next revision
draft [2009/07/15 15:49]
ptacek
draft [2009/07/15 15:59]
ptacek
Line 14: Line 14:
 NLP server s tectomt, ASR/TTS/SR client, connected over network NLP server s tectomt, ASR/TTS/SR client, connected over network
 XXX JPta XXX JPta
- 
-advances in Czech NLU (on reconstructed spoken data): 300-500vet(?) rucne anotovat pos, a-tree, t-tree, IE predicates, Named Entities, DA pro eval in-domain testy after Nov. 
-pos ? analyzovat, generovat a kontrolovat 'jen' kde je rozdil ve forme? 
- 
- 
  
  
Line 77: Line 72:
 features: domain tailored tagset (variation of DAMSL-SWBD) features: domain tailored tagset (variation of DAMSL-SWBD)
 performance indicator: accuracy performance indicator: accuracy
 +
 +
 +
  
  
Line 83: Line 81:
 ===== Dialog Manager (WP 5.3) ===== ===== Dialog Manager (WP 5.3) =====
 features: reply types, using (language independed) predicates (prakticky to znamena, ze pojmenuju testy na prechodech v dafech anglicky) features: reply types, using (language independed) predicates (prakticky to znamena, ze pojmenuju testy na prechodech v dafech anglicky)
-Manually created DAFs covering following topics: Person_Retired, Person_in_productive_age, Child, Husband, Wife, Wedding, Death, Christmas, Handle_stalled_dialog +Manually created DAFs covering following topics: Person_Retired, Person_in_productive_age, Child, Husband, Wife, Wedding, Death, Christmas, Handle_stalled_dialog (most frequent topics in corpora) 
-performance indicator: rucni hodnoceni prijatelnosti vybrane akce+performance indicator: acceptability - manual evaluation of actions selected by DM 
 + 
  
  
 ===== Natural Language Generation (WP 5.4) ===== ===== Natural Language Generation (WP 5.4) =====
-features: paraphrases for hard-coded utterances, underspecified input (dott format), passing-through emotional markup (natvrdo v dafech a templatech u hodnoticich vet)+features: morphological adjustments, generating paraphrases for hard-coded utterances, underspecified input (dott format), passing-through emotional markup (natvrdo v dafech a templatech u hodnoticich vet)
 performance indicator: BLEU score performance indicator: BLEU score
  
Line 106: Line 106:
 T5.2.7 tohle zminuje, nick webb to pro nas asi neudela T5.2.7 tohle zminuje, nick webb to pro nas asi neudela
 performance indicator: number of tokens in user reply utterances, post-session questionare performance indicator: number of tokens in user reply utterances, post-session questionare
 +
 +
 +===== advances =====
 +
 +advances in Czech NLU (on reconstructed spoken data): 300-500vet(?) rucne anotovat pos, a-tree, t-tree, IE predicates, Named Entities, DA pro eval in-domain testy after Nov.
 +pos ? analyzovat, generovat a kontrolovat 'jen' kde je rozdil ve forme?

[ Back to the navigation ] [ Back to the content ]