Báo cáo khoa học: "Phrase-based Statistical Language Generation using Graphical Models and Active Learning" potx

... LinguisticsPhrase-based Statistical Language Generation using Graphical Models and Active LearningFrançois Mairesse, Milica Gaˇsić, Filip Jurˇc´ıˇcek,Simon Keizer, Blaise Thomson, Kai Yu and Steve ... E. Schapire, and D. Hakkani-Tür. Active learn-ing for spoken language understanding. In Proceedings ofICASSP, 2003.S. Varges and C. Mellish. Instance-based natural language generation. In ... sizes, using random sampling and active learning. Differences for training setsizes of 20 and 40 are all significant (p < .05).terestingly, while models trained on 100 utterancesoutperform models...

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

... M., and Popovici C. 1997. Contextual Information and Spe-cific Language Models for Spoken Language Un-derstanding. In Proceedings of SPECOM’97, Cluj-Napoca, Romania, pp. 51–56.Bangalore S. and ... movespecific statistical language models (DM-SLMs)by using GF to generate all utterances that arespecific to certain dialogue moves from our in-terpretation grammar. In this way we can pro-duce models ... Hypotheses Reordering Using Linear Re-gression. In Proceedings of Eurospeech 2001. Aal-borg, Denmark, pp. 1829–1832.Clarkson P.R. and Rosenfeld R. 1997. Statistical Language Modeling Using the CMU-CambridgeToolkit....

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

... phonological models were de-signed to represent non-CP ASL, but CPs use a reduced set of handshapes, standard eye-gaze and head-tilt patterns, and more complex orientations and motion paths. ... 2004a, 2004b, 2005), focusing on ASL generation. This overview illustrates important correspondences between the problem of ASL natural language generation (NLG) and related research in Multimodal ... applications for the generation of gesture animation and other communication signals that are not easily encoded as text strings. 1 Introduction and Motivations American Sign Language (ASL) is...

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

... Integrating Natural Language Generation with XML Web TechnologyGraham WilcockUniversity of Helsinki00014 Helsinki, Finlandgraham.wilcock@helsinki.fiAbstractThe paper ... for natural language gen-eration are described by Wilcock (2001), based onpractical experience in developing an XML-based generation component for a spoken dialogue sys-tem (Jokinen and Wilcock, ... byReiter and Dale (2000) has the following modules and tasks:•Text Planning—content determination—discourse structuring•Microplanning—lexicalization—referring expression generation —aggregation•Realization—linguistic...

Báo cáo khoa học: "Boosting Statistical Word Alignment Using Labeled and Unlabeled Data" ppt

... Computational LinguisticsBoosting Statistical Word Alignment Using Labeled and Unlabeled Data Hua Wu Haifeng Wang Zhanyi Liu Toshiba (China) Research and Development Center 5/F., Tower ... word aligner by using both the labeled data and the unlabeled data. Then we build a pseudo reference set for the unlabeled data, and calculate the error rate of each word aligner using only ... bilingual corpora (Wu, 1997; Och and Ney, 2003; Cherry and Lin, 2003; Wu et al., 2005; Zhang and Gildea, 2005). These methods unsupervisedly train the alignment models with unlabeled data. A...

Báo cáo khoa học: "A NATUWAL LANGUAGE INTERFACE USING A WORLD MODEL" pdf

... the fields of housing, sales, and drug testing, thus confirming its transportability and practicality. INTRODUCTION KID (Knowledge-based Interface to Databases) is a Japanese -language database ... Extendability and robustness Natural language sentences employ a wide variety of expressions. A parser must always be extended to understand new sentences. A parser which can understand one set ... Japanese -language sentences), the model- based parser, semantic interpretation and the flow of the parsing process, knowledge for customizing KID and, lastly, the evaluation of KID and its...

Báo cáo khoa học: "Subjective Natural Language Problems: Motivations, Applications, Characterizations, and Implications" pptx

... insightsthat may lead to a more thorough understanding of language uses and users. Such insights may in factdecisively advance language science and artificialnatural language intelligence.AcknowledgmentsI ... interfac-ing (Liu et al, 2003; Holzman and Pottenger, 2003;Francisco and Gervás, 2006; Kalra and Karahalios,2005; Généreux and Evans, 2006; Mihalcea and Liu,2006). Affective semantics ... into sociolin-guistic and interactional insights on language (Mesthrie et al, 2009).• Multimodality and interdisciplinarity Manyof these problems have an interactive and hu-manistic basis....

Báo cáo khoa học: "Robust PCFG-Based Generation using Automatically Acquired LFG Approximations" doc

... features.Valldal and Oepen (2005) present a discriminativedisambiguation model using a hand-crafted HPSGgrammar for generation. Belz (2005) describesa method for building statistical generation mod-els using ... to rank output strings and ad-ditional statistical and symbolic resources at in-termediate generation stages. Ratnaparkhi (2000)uses maximum entropy models to drive generation with word bigram ... and not easily ported to new domains orother languages. Our methodology, on the otherhand, is based on resources automatically acquiredfrom treebanks and easily ported to new domains and languages,...

Báo cáo khoa học: "A Stochastic Language Model using Dependency and Its Improvement by Word Clustering" ppt

... two complete stochastic language models using dependency between bugsetsu, a sequence of one or more content words followed by zero, one or more function words, and evaluate their predictive ... bunsetsu, the content word sequence and the function word sequence are independently predicted by word n-gram models equipped with unknown word models (Mori and Yamaji, 1997). The above model ... G1 and G2, whose 900 probabilities axe P1 and P2 respectively, is repre- sented as follows: P(A =~ a) = ~IPI(A =~ c~) +,~P2(A =~ a) 0<~j < l(j=l, 2) and ~,+~2=1 (4) where A E V and...

Tài liệu Báo cáo khoa học: "Modeling Morphologically Rich Languages Using Split Words and Unstructured Dependencies" docx

... speechrecognition and machine translation (Bahl et al.,1983; Brown et al., 1990). More recently, good re-sults on lexical substitution and word sense disam-biguation using language models have also ... 2gives the total log-probability (using log2) for thesplit and unsplit datasets using n-gram models of different order. We compute the perplexityof the two datasets using a common denomina-tor: ... structured lan-guage models (Chelba and Jelinek, 2000), and de-pendency language models (Chelba et al., 1997).The dependency model governs the whole sen-tence and each word in a sentence is likely...

Xem thêm