Matches in SemOpenAlex for { <https://semopenalex.org/work/W2893411096> ?p ?o ?g. }
Showing items 1 to 90 of
90
with 100 items per page.
- W2893411096 abstract "Spoken language understanding (SLU) systems are widely used in handling of customer-care calls.A traditional SLU system consists of an acoustic model (AM) and a language model (LM) that areused to decode the utterance and a natural language understanding (NLU) model that predicts theintent. While AM can be shared across different domains, LM and NLU models need to be trainedspecifically for every new task. However, preparing enough data to train these models is prohibitivelyexpensive. In this paper, we introduce an efficient method to expand the limited in-domain data. Theprocess starts with training a preliminary NLU model based on logistic regression on the in-domaindata. Since the features are based onn= 1,2-grams, we can detect the most informative n-gramsfor each intent class. Using these n-grams, we find the samples in the out-of-domain corpus that1) contain the desired n-gram and/or 2) have similar intent label. The ones which meet the firstconstraint are used to train a new LM model and the ones that meet both constraints are used to train anew NLU model. Our results on two divergent experimental setups show that the proposed approachreduces by 30% the absolute classification error rate (CER) comparing to the preliminary modelsand it significantly outperforms the traditional data expansion algorithms such as the ones based onsemi-supervised learning, TF-IDF and embedding vectors." @default.
- W2893411096 created "2018-10-05" @default.
- W2893411096 creator A5060308914 @default.
- W2893411096 creator A5061105309 @default.
- W2893411096 creator A5091420288 @default.
- W2893411096 date "2018-09-27" @default.
- W2893411096 modified "2023-09-27" @default.
- W2893411096 title "Automatic Data Expansion for Customer-care Spoken Language Understanding" @default.
- W2893411096 cites W1578856370 @default.
- W2893411096 cites W1978394996 @default.
- W2893411096 cites W2095734449 @default.
- W2893411096 cites W2117130368 @default.
- W2893411096 cites W2117278770 @default.
- W2893411096 cites W2149980590 @default.
- W2893411096 cites W2158195707 @default.
- W2893411096 cites W2184045248 @default.
- W2893411096 cites W2210838531 @default.
- W2893411096 cites W2408500895 @default.
- W2893411096 cites W2474824677 @default.
- W2893411096 cites W2891229414 @default.
- W2893411096 cites W2950133940 @default.
- W2893411096 cites W2963042536 @default.
- W2893411096 hasPublicationYear "2018" @default.
- W2893411096 type Work @default.
- W2893411096 sameAs 2893411096 @default.
- W2893411096 citedByCount "4" @default.
- W2893411096 countsByYear W28934110962019 @default.
- W2893411096 countsByYear W28934110962021 @default.
- W2893411096 countsByYear W28934110962022 @default.
- W2893411096 crossrefType "posted-content" @default.
- W2893411096 hasAuthorship W2893411096A5060308914 @default.
- W2893411096 hasAuthorship W2893411096A5061105309 @default.
- W2893411096 hasAuthorship W2893411096A5091420288 @default.
- W2893411096 hasConcept C119857082 @default.
- W2893411096 hasConcept C134306372 @default.
- W2893411096 hasConcept C137293760 @default.
- W2893411096 hasConcept C154945302 @default.
- W2893411096 hasConcept C162324750 @default.
- W2893411096 hasConcept C187736073 @default.
- W2893411096 hasConcept C195324797 @default.
- W2893411096 hasConcept C204321447 @default.
- W2893411096 hasConcept C2775852435 @default.
- W2893411096 hasConcept C2776230583 @default.
- W2893411096 hasConcept C2779439875 @default.
- W2893411096 hasConcept C2780451532 @default.
- W2893411096 hasConcept C33923547 @default.
- W2893411096 hasConcept C36503486 @default.
- W2893411096 hasConcept C41008148 @default.
- W2893411096 hasConceptScore W2893411096C119857082 @default.
- W2893411096 hasConceptScore W2893411096C134306372 @default.
- W2893411096 hasConceptScore W2893411096C137293760 @default.
- W2893411096 hasConceptScore W2893411096C154945302 @default.
- W2893411096 hasConceptScore W2893411096C162324750 @default.
- W2893411096 hasConceptScore W2893411096C187736073 @default.
- W2893411096 hasConceptScore W2893411096C195324797 @default.
- W2893411096 hasConceptScore W2893411096C204321447 @default.
- W2893411096 hasConceptScore W2893411096C2775852435 @default.
- W2893411096 hasConceptScore W2893411096C2776230583 @default.
- W2893411096 hasConceptScore W2893411096C2779439875 @default.
- W2893411096 hasConceptScore W2893411096C2780451532 @default.
- W2893411096 hasConceptScore W2893411096C33923547 @default.
- W2893411096 hasConceptScore W2893411096C36503486 @default.
- W2893411096 hasConceptScore W2893411096C41008148 @default.
- W2893411096 hasLocation W28934110961 @default.
- W2893411096 hasOpenAccess W2893411096 @default.
- W2893411096 hasPrimaryLocation W28934110961 @default.
- W2893411096 hasRelatedWork W1576375683 @default.
- W2893411096 hasRelatedWork W1912606886 @default.
- W2893411096 hasRelatedWork W2000577984 @default.
- W2893411096 hasRelatedWork W2147262247 @default.
- W2893411096 hasRelatedWork W2578330760 @default.
- W2893411096 hasRelatedWork W2588225431 @default.
- W2893411096 hasRelatedWork W2735125076 @default.
- W2893411096 hasRelatedWork W2743354079 @default.
- W2893411096 hasRelatedWork W2939335894 @default.
- W2893411096 hasRelatedWork W2950396480 @default.
- W2893411096 hasRelatedWork W2955579501 @default.
- W2893411096 hasRelatedWork W2962782699 @default.
- W2893411096 hasRelatedWork W3029923195 @default.
- W2893411096 hasRelatedWork W3090635823 @default.
- W2893411096 hasRelatedWork W3103090120 @default.
- W2893411096 hasRelatedWork W3132773134 @default.
- W2893411096 hasRelatedWork W3173042348 @default.
- W2893411096 hasRelatedWork W3179305276 @default.
- W2893411096 hasRelatedWork W3199922048 @default.
- W2893411096 hasRelatedWork W3207345702 @default.
- W2893411096 isParatext "false" @default.
- W2893411096 isRetracted "false" @default.
- W2893411096 magId "2893411096" @default.
- W2893411096 workType "article" @default.