[ Skip to the content ]

Institute of Formal and Applied Linguistics Wiki

[ Back to the navigation ]

You are here: start » draft

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 14:58]
ptacek 		draft [2009/07/15 15:49]
ptacek
Line 26: Line 26:
 features: improved language models, real-time speaker adaptation features: improved language models, real-time speaker adaptation
 performance indicator: WER performance indicator: WER
 +
  
  
Line 32: Line 33:
  
 ===== Speech Reconstruction (WP 5.1 ???) ===== ===== Speech Reconstruction (WP 5.1 ???) =====
-features: omit filler phrases, irrelevant speech events, false starts, repetitions, corrections, syntax smoothing (word ordering, +features: omit filler phrases, remove irrelevant speech events, handle false starts, repetitions, and corrections, polish word ordering
 performance indicator: BLEU score between actual output and manually reconstructed sentences from corpora (T5.2.1), baseline: Moses with default settings performance indicator: BLEU score between actual output and manually reconstructed sentences from corpora (T5.2.1), baseline: Moses with default settings
 +
 +
 +
 +
 +
  
 ===== Morphology Analyzer and POS tagging (WP 5.2) ===== ===== Morphology Analyzer and POS tagging (WP 5.2) =====
-features: XXX Mirek/Johanka +features: coverage of photo-pal domain, domain adapted tagger (XXX prida nam Jarka OOV slova co najdeme, bude PDTSC rucne oznackovane - do listopadu?) 
-performance indicator: accuracy+performance indicator: OOV rate, accuracy 
 + 
  
 ===== Syntactic Parsing (WP 5.2) ===== ===== Syntactic Parsing (WP 5.2) =====
 features: induce dependencies and labels features: induce dependencies and labels
-performance indicator: f-measure+performance indicator: accuracy (correctly induced edges, labels)
 v tipu je natrenovat MacDonnalda na dialog datech, ten task je do M42, ted ne. v tipu je natrenovat MacDonnalda na dialog datech, ten task je do M42, ted ne.
 +
 +
  
  
  
 ===== Semantic Parsing (WP 5.2) ===== ===== Semantic Parsing (WP 5.2) =====
-features: meaning representation with semantic roles (69 roles), coordinations, argument structure, partial ellipsis resolution, pronominal anaphora resolution, +features: assignment of semantic roles (69 roles), coordinations, argument structure, partial ellipsis resolution, pronominal anaphora resolution, post parsing detection of ungrammatical edges (caused by long utterances) 
-performance indicator: f-measure+performance indicator: accuracy (correctly induced edges, labels) 
  
 ===== Information Extraction (WP 5.2) ===== ===== Information Extraction (WP 5.2) =====
Line 54: Line 64:
 covering predicates from  before-mentioned set of DAFs. covering predicates from  before-mentioned set of DAFs.
 performance indicator: accuracy performance indicator: accuracy
 +
  
  
 ===== Named Entities Recognition (WP 5.2) ===== ===== Named Entities Recognition (WP 5.2) =====
-features: detect person names, geographical locations (organizations myslim nepotrebne)+features: detect person names, geographical locations, organizations
 performance indicator: f-measure performance indicator: f-measure
 +
 +
 +
  
 ===== Dialog Act Tagging (WP 5.2) ===== ===== Dialog Act Tagging (WP 5.2) =====
-features: tagset derived from DAMSL-SWBD, DA is a key feature driving the decision, what to say next.+features: domain tailored tagset (variation of DAMSL-SWBD)
 performance indicator: accuracy performance indicator: accuracy
  
  
-===== Sentiment Analysis (WP 5.2) ===== 
-features: za tohle bych vydaval klasifikator, co rozhoduje ,jestli se rekne 'To je smutné/veselé'. Tem adjektivum rucne priradim negative/positive sentiment. 
-performance indicator: f-measure 
  
  
 +===== Dialog Manager (WP 5.3) =====
 +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
 +performance indicator: rucni hodnoceni prijatelnosti vybrane akce
  
  
 +===== 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)
 +performance indicator: BLEU score
  
  
-===== Complete System Evaluation ===== +====== AZ PO LISTOPADU ======
-T5.2.7 tohle zminuje, nick webb to pro nas asi neudela +
-performance indicator: number of tokens in user reply utterances, post-session questionare+
  
 +===== Syntactic Parsing (WP 5.2) =====
 +features: adapted to domain (McD trained on manual PDTSC trees)
 +performance indicator: accuracy (correctly induced edges, labels)
  
  
 +===== Sentiment Analysis (WP 5.2) =====
 +features: za tohle bych vydaval klasifikator, co rozhoduje ,jestli se rekne 'To je smutné/veselé'. Tem adjektivum rucne priradim negative/positive sentiment.
 +performance indicator: f-measure
  
-===== Dialog Manager (WP 5.3) ===== +===== Complete System Evaluation ===== 
-features: reply types, using (language independed) predicates (prakticky to znamena, ze pojmenuju testy na prechodech v dafech anglicky) +T5.2.7 tohle zminuje, nick webb to pro nas asi neudela 
-Handmade DAF covering following topics: Person_Retired, Person_in_productive_age, Child, Husband, Wife, Wedding, Christmas, Handle_stalled_dialog +performance indicator: number of tokens in user reply utterances, post-session questionare
-performance indicator: rucni hodnoceni prijatelnosti vybrane akce +
- +
-===== Natural Language Generation (WP 5.4) ===== +
-features: variations, underspecified input (dott format), emotional markup (natvrdo v dafech a templatech u hodnoticich vet) +
-performance indicator: BLEU score+

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