Báo cáo khoa học: "A Fast and Accurate Method for Approximate String Search" pptx

... Association for Computational Linguistics, pages 52–61,Portland, Oregon, June 19-24, 2011.c2011 Association for Computational LinguisticsA Fast and Accurate Method for Approximate String SearchZiqi ... consists ofcandidate generation and candidate ﬁnal selection.The former task is an example of approximate string search. Note that candidate generation is only con-cerned with a single word. For single-word ... to as approximate string search. Given aquery string, a dictionary of strings (vocabulary), and a set of operators, the system returns the topk strings in the dictionary that can be transformedfrom...

Tài liệu Báo cáo khoa học: "A Syntax-Driven Bracketing Model for Phrase-Based Translation" pptx

... Experimental results and analy-sis show that the new model outperformsother previous methods and achieves asubstantial improvement over the baselinewhich is not syntactically informed.1 IntroductionThe ... NP”, and “VP→ADVPVP” are used as features for s1, s2 and srespectively.• Path Features (PF)The tree path σ(s1) σ(s) connecting σ(s1) and σ(s), σ(s2) σ(s) connecting σ(s2) and σ(s), ... Marton and Resnik use for Chinese-to-English translation. Our experimental results dis-play that our SDB model achieves a substantialimprovement over the baseline and signiﬁcantlyoutperforms...

Tài liệu Báo cáo khoa học: "A Hybrid Convolution Tree Kernel for Semantic Role Labeling" pptx

... 2005b) and post inference (Pun-yakanok et al., 2004).These feature-based methods are considered asthe state of the art method for SRL and achievedmuch success. However, as we know, the standardﬂat ... kernels, more and morekernels for restricted syntaxes or speciﬁc do-mains, such as string kernel for text categoriza-tion (Lodhi et al., 2002), tree kernel for syntacticparsing (Collins and Duffy, ... convo-lution tree kernel for SRL. In this paper, we fol-low the same framework and design a novel hybridconvolution kernel for SRL.3 Feature-based methods for SRLUsually feature-based methods refer...

Báo cáo khoa học: "A Latent Dirichlet Allocation method for Selectional Preferences" pptx

... Association for Computa-tional Linguistics.Andrew Carlson, Justin Betteridge, Richard C. Wang,Estevam R. Hruschka Jr., and Tom M. Mitchell.2010. Coupled semi-supervised learning for infor-mation ... beretrained and tuned for each task. On the otherhand, generative models produce complete prob-ability distributions of the data, and hence can beintegrated with other systems and tasks in ... predicate and argument, and evaluation is performed on aset of human-plausibility judgments obtaining im-pressive results against Keller and Lapata’s (2003)Web hit-count based system. Van Durme and Gildea...

Báo cáo khoa học: "A Morphological Analysis Based Method for Spelling Correction" docx

...

Báo cáo khoa học: "A Generalized-Zero-Preserving Method for Compact Encoding of Concept Lattices" pot

... or join of u, v ∈ X, if one exists, and u  v for the greatest lower bound or meet. If weneed a second partial order, we use  for its orderrelation and  for its join operation. We are espe-cially ... encodings that use sets with intersection for meet and empty set for failure: with Y deﬁnedas the power set of some set Z,  as ⊇,  as ∩, and g(f(u), f(v)) = 0 if and only if f(u) ∩ f(v) = ∅,then ... spaces and its application to typed feature struc-tures. In Proceedings of the 40th Annual Meeting ofthe Association for Computational Linguistics (ACL2002), pages 64–71.Andrew Sadler and Carmen...

Báo cáo khoa học: "A Hierarchical Phrase-Based Model for Statistical Machine Translation" pptx

... designating the source and target languages as “French” and “English,” respectively. The variables f and e stand for source and target sentences; fjistands for the substring of ffrom position ... standing for X spanning fji. We choose b and β to balancespeed and performance on our development set. For our experiments, we set b = 40, β = 10−1 for X cells, and b = 15, β = 10−1 for ... ek for some k ∈ [i, j] and k∈ [i, j];2. fk ek for all k ∈ [i, j] and k [i, j];3. fk ek for all k  [i, j] and k∈ [i, j].Next, we form all possible diﬀerences...

Báo cáo khoa học: "A MARKOV LANGUAGE LEARNING MODEL FOR FINITE PARAMETER SPACES" pptx

... phrase, and so forth. Given these alternatives, Gibson and Wexler (1994) show that there are 12 possible surface strings for each (-V2) grammar and 18 possible surface strings for each (+V2) ... ACQUISITION Recently, several researchers, including Gibson and Wexler (1994), henceforth GW, Dresher and Kaye (1990); and Clark and Roberts (1993) have modeled language learning in a (finite) ... "degree-0" examples for reasons of psychological plausibility (see Gibson and Wexler for discussion). For in- stance, the parameter setting [0 1 0]= Specifier initial, Complement final, and -V2, works...

Tài liệu Báo cáo khoa học: "A Fast, Accurate Deterministic Parser for Chinese" pdf

... S(1) and S(2) or not.4 Last action given by the classifier, and number of words in S(1) and S(2).5 Headword and its POS of S(1), S(2), S(3) and S(4), and word and POS of Q(1), Q(2), Q(3) and ... the p-value for LR and LP are 0.008 and 0.457, respectively. The increase in recallis statistically significant, and it shows classifierstacking can improve performance.On the other hand, we did ... set) for training and 10% for testing. Funget al.(2004) also took the maximum-entropy mod-eling approach, but augmented by transformation-based learning. They used the standard trainingand...

Tài liệu Báo cáo khoa học: "Semantic Information and Derivation Rules for Robust Dialogue Act Detection in a Spoken Dialogue System" pptx

... PST as a knowl-edge source for DA detection, we essentially need tomodel the relationship between the random DA and the random DR. Denote the random DA by X and the random DR by Y . Given a text ... z-score (Larsen and Marx,2000) of each word w in the ASR output. Figure 3illustrates the PST for the sentences: Where is theAnping-Fort. There are two keywords Where and Anping-Fort and two non-keywords ... Association for Computational Linguistics:shortpapers, pages 603–608,Portland, Oregon, June 19-24, 2011.c2011 Association for Computational LinguisticsSemantic Information and Derivation Rules for...

Xem thêm