Báo cáo khoa học: "American Sign Language Generation: Multimodal NLG with Multiple Linguistic Channels" pot

... Michigan, June 2005.c2005 Association for Computational LinguisticsAmerican Sign Language Generation: Multimodal NLG with Multiple Linguistic Channels Matt Huenerfauth Computer and Information ... for a deaf user. 2 ASL NLG: A Form of Multimodal NLG NLG researchers think of communication signals in a variety of ways: some as a written text, other as speech audio (with prosody, timing, ... most Multimodal NLG systems, which have a single linguistic channel/modality that is coordinated with other non -linguistic resources (Figure 1). Figure 1: Linguistic Channels in Multimodal Systems...

Báo cáo khoa học: "Integrating Natural Language Generation with XML Web Technology" doc

... paper describes a software demointegrating Natural Language Genera-tion (NLG) techniques with recent de-velopments in XML web technology.The NLG techniques include a form oftemplate-based generation, ... software demo basedon this approach, showing a concrete example.2 Brief Summary of NLG The most common design for NLG systems is apipeline architecture. The version described byReiter and Dale ... Integrating Natural Language Generation with XML Web TechnologyGraham WilcockUniversity of Helsinki00014 Helsinki, Finlandgraham.wilcock@helsinki.fiAbstractThe...

Tài liệu Báo cáo khoa học: "Generating statistical language models from interpretation grammars in dialogue systems" potx

... with a comparison of in-grammar recognition performance.3 Language modellingTo generate the different trigram language modelswe used the SRI language modelling toolkit (Stol-cke, 2002) with ... thesame language as the Nuance grammar, gives suchan important gain in performance that it lowersthe WER with 22%. We used the Chi-square testof signiﬁcance to statistically compare the results with ... modelsinterpolated with the GSLC corpus and the domainnews corpus in different ways which at best givesa relative reduction in WER of 8% in comparison with MP3GFLM and 43% compared with the base-line....

Báo cáo khoa học: "A Cross-Language Document Retrieval System Based on Semantic Annotation" pot

... domains,given appropriate linguistic and semanticresources.1 IntroductionThe task of a cross -language information re-trieval (CUR) system is to match user queriesspecified in one language against ... information that the retrieval system is us-ing. Crossing the language barrier from a queryin one language to the document collection inanother language is done via concept codes as aninterlingua representation. ... documents into an intermediary1 The Unified Medical Language System(http://umls.nlm.nih.gov/research/umls/) integrates informa-tion from multiple machine-readable biomedical informationsources.2...

Tài liệu Báo cáo khoa học: "Extending the Entity-based Coherence Model with Multiple Ranks" pot

... Computational Linguistics, pages 315–324,Avignon, France, April 23 - 27 2012.c2012 Association for Computational LinguisticsExtending the Entity-based Coherence Model with Multiple RanksVanessa ... experiment with thestratified option, in which C ranks are assigned tothe permutations generated from the same sourcedocument. The permutation with the smallest dis-similarity score is assigned ... (2008).In sentence ordering, we experiment with different approaches to assigning dissimilarityscores and ranks (Section 5.1.1). We also exper-iment with different entity extraction approaches315Manila...

Báo cáo khóa học: Damped oscillatory hysteretic behaviour of butyrylcholinesterase with benzoylcholine as substrate potx

... showed that the inductiontime, s, with BzCh as substrate was much shorter than with NMIA [16]. Moreover, with NMIA and BuSCh, it wasfound that s increased with decreasing pH [25]. Therefore,to ... Indeed, with BuSCh, deacylation is partly ratelimiting for wild-type and fully rate-limiting for D70G [37]. With BzSCh, kcatis much lower than kcatfor BzCh (18-fold with wild-type and 12-fold with ... mechanical calculations should verify ordisprove these contentions.Hysteresis with BzCh but not with BzSChAs with NMIA, BuChE-catalyzed hydrolysis of BzCh wasfound to present a slow hysteretic...

Báo cáo khoa học: "Improving the Interpretation of Noun Phrases with Cross-linguistic Information" doc

... Computational Linguistics, pages 568–575,Prague, Czech Republic, June 2007.c2007 Association for Computational LinguisticsImproving the Interpretation of Noun Phrases with Cross -linguistic InformationRoxana ... constructions ofthe form N P N in Romance languages where, aswe will show below, the P (preposition) varies inways that correlate with the semantics. Thus Ro-mance languages will give us another source ... N N and N P N in all languages consideredand resulted after agreement.4 Model4.1 Feature spaceWe have identiﬁed and experimented with 13 NPfeatures presented below. With the exceptions offeatures...

Tài liệu Báo cáo khoa học: "How spoken language corpora can reﬁne current speech motor training methodologies" pptx

... onproduction accuracy in experiments with speakers with AOS and speakers with conduction aphasia.3.2 Implications on motor learningIn that view, practicing with high-frequency sylla-bles could ... community. Existing applications forselecting linguistic stimuli, although undoubtedlyuseful, are not based on spoken language data,and are generally not designed for utilization byspeech therapists ... decisive role with re-spect to articulatory accuracy in the spontaneousspeech production of patients with AOS (Staigeret al., 2008). Similarly, (Laganaro, 2008) con-ﬁrms a signiﬁcant effect...

Tài liệu Báo cáo khoa học: "Incremental Syntactic Language Models for Phrase-based Translation" pptx

... constructs new hypotheses by extend-ing existing hypotheses. New hypotheses are placedin appropriate hypothesis stacks. In the simplestcase, a new hypothesis extends an existing hypothe-sis by ... phrase-based translation lat-tice. Within each decoder stack t, each hypothe-sis h is augmented with a syntactic language modelstate ˜τth. Each syntactic language model state isa random variable ... Based Translation with anIncremental Syntactic Language ModelThe phrase-based decoder is augmented by addingadditional state data to each hypothesis in the de-626coder’s hypothesis stacks....

Tài liệu Báo cáo khoa học: "The Natural Language Toolkit" docx

... Viterbi-style PCFGparser, together with a suite of bottom-upprobabilistic chart parsers.4 Teaching with NLTKNatural language processing is often taught withinthe conﬁnes of a single-semester ... USAAbstractThe Natural Language Toolkit is a suite ofprogram modules, data sets and tutorialssupporting research and teaching in com-putational linguistics and natural language processing. NLTK ... be parsed using a grammar with left-recursive productions, so long as weuse a chart parser. We can invoke NLTK’s chartparser with a bottom-up rule-invocation strategy with chart.ChartParse(grammar,chart.BUSTRATEGY)....

Xem thêm