Báo cáo khoa học: "A Hierarchical Bayesian Language Model based on Pitman-Yor Processes" docx

... machinetranslation. Most language models fall into theclass of n-gram models, which approximate thedistribution over sentences using the conditionaldistribution of each word given a context consist-ing ... novel language model based on a hierarchical Bayesian model (Gelman et al., 1995) where each hidden variableis distributed according to a Pitman-Yor process, anonparametric generalization of ... Process Pitman-Yor processes are examples of nonpara-metric Bayesian models. Here we give a quick de-scription of the Pitman-Yor process in the contextof a unigram language model; good tutorials on such...

Báo cáo khoa học: "A FLEXIBLE NATURAL LANGUAGE PARSER BASED ON A TWO-LEVEL REPRESENTATION OF SYNTAX" ppt

... Verbs, copulas REF Referent Nouns, pronouns CONN Connector Prepositions, conjunctions Articles, DET Determiner demonstrative adjectives, adjectival question words Adverbial MOD Adverbs Modifier ... other condi tions evaluate to false), its action part is exe cured. An action consists in the construction of new nodes, in their filling up with particular val ues (normally depending on the ... conditions are evaluated. The conditions involve tests on the cur rent structure of the tree (i.e. the "status" of the analysis) and may request a one-word lookahead. If just one...

Báo cáo khoa học: "Automatic Acquisition of Language Model based on Head-Dependent Relation between Words" pdf

... grammar, grammar -based language modeling is expected to be more powerful and compact in model size than n-gram -based one. In this paper we present a language modeling based on a kind of simple ... statistical machine translation. Previous works for language modeling can be broadly divided into two approaches; one is n- gram -based and the other is grammar -based. N-gram model estimates the probability ... statistical machine translation. In this paper, we present a language modeling based on a kind of simple dependency grammar. The grammar consists of head-dependent relations between words and can...

... novel FHMM -based pro-noun resolution system.2 Model DescriptionThis work is based on a graphical model frameworkcalled Factorial Hidden Markov Models (FHMMs).Unlike the more commonly known ... few considerations made us reconsider.First, exceptions are found in the corpus. Personalpronouns such as she or he are used to refer to coun-try, regions, states or organizations. Second, ... and B. Wellner. 2003. Toward cond i-tional models of identity uncertainty with applicationto proper noun coreference. In IJCAI Workshop on In-formation Integration on the Web. Citeseer.David...

Báo cáo khoa học: "a Chat-oriented Dialogue System based on the Vector Space Model" ppt

... This system demonstration paper presents IRIS (Informal Response Interactive Sys-tem), a chat-oriented dialogue system based on the vector space model framework. The system belongs to the class ... implementation. Section 3 presents some illustrative examples of dialogues generated by IRIS, and Section 4 presents the main conclusions of this work. 372 The IRIS Implementation In this section ... mentioned, IRIS architecture is heavily based on a vector space model framework, which includes a standard similarity search module from vector -based information retrieval systems (Salton and...

Báo cáo khoa học: "A Generative Constituent-Context Model for Improved Grammar Induction" docx

... fromthe conditional model used for induction. Here, weimprove on that model in several ways. First, weconstruct a generative model which utilizes the samefeatures. Then, we extend the model to ... manning}@cs.stanford.eduAbstractWe present a generative distributional model for theunsupervised induction of natural language syntaxwhich explicitly models constituent yields and con-texts. Parameter search with EM produces ... funda-mental assumption is a much weakened version ofclassic linguistic constituency tests (Radford, 1988):constituents appear in constituent contexts. A par-ticular linguistic phenomenon that the...

Tài liệu Báo cáo khoa học: "A Fully Bayesian Approach to Unsupervised Part-of-Speech Tagging∗" docx

... pre-senting results for conditions ranging from a full tagdictionary to no dictionary at all. We introduce theuse of a new information-theoretic criterion, varia-tion of information (Meilˇa, 2002), ... a derivation.2.2 Model DeﬁnitionOur model has the structure of a standard trigramHMM, with the addition of symmetric Dirichlet pri-ors over the transition and output distributions:ti|ti−1= ... The only complication is that resamplinga tag changes the identity of three trigrams at once,and we must account for this in computing its condi-tional distribution. The sampling distribution...

Tài liệu Báo cáo khoa học: "A Hierarchical Account of Referential Accessibility" pptx

... bookshelves tonight.Figure 8: RST tree for text (6), using rhetoricalrelationsFigure 9: RST tree for text (6), using intention -based relationsConclusionVeins Theory is based on established notions ... structure: hierarchical organization, asin the stack -based model and RST's treestructures, and dominance or nuclear/satellitemotivationB6-C6motivationB6C6A6A6-B6conditionconditionA6B6C6 ... VTperforming consistently better than the stack- based model. Figure 6: Effort required by VT and stack -based modelsNote that in some cases, the stack -based model performs better than VT, in particular...

Báo cáo khoa học: "A Nonparametric Bayesian Approach to Acoustic Model Discovery" docx

... acoustic model with three supervised ones: 1) an English triphone model, 2) an English monophone model and 3) aThai monophone model. The ﬁrst two were trained on TIMIT, while the Thai monophone model ... sub-word models to guideits hypotheses on phone boundaries. Bayesian Model for Segmentation Our model isinspired by previous applications of nonparametric Bayesian models to segmentation problems ... evaluation metrics for our model allconverged. In Section 7, we report the performanceof our model using the sample from the last iteration.7 ResultsFig. 4 shows a confusion matrix of the 48 phonesused...

Báo cáo khoa học: "A Hierarchical Model of Web Summaries" docx

... hierar-chical Bayesian model, closely followed by theinformation-theoretical model using entropy as itsselection criterion.6 ConclusionIn this paper we have demonstrated the creation of atopic -model ... entropy- based model when restricted to the non-regionalsection of the corpus. Also if the reduction inperplexity remains limited for the entropy, χ2andhLLDA models, the cross-entropy based model ... Reference ModelsOur reference models consists of several n-gram(n ∈ [1, 3]) language models, none of which makesuse of the hierarchical information available fromthe corpus. Under these models,...

Xem thêm