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 Linguistics A Fast and Accurate Method for Approximate String Search Ziqi ... consists of candidate generation and candidate final selection. The former task is an example of approximate string search. Note that candidate generation is only c...

Ngày tải lên: 30/03/2014, 21:20

... Experimental results and analy- sis show that the new model outperforms other previous methods and achieves a substantial improvement over the baseline which is not syntactically informed. 1 Introduction The ... NP”, and “VP→ADVP VP” are used as features for s 1 , s 2 and s respectively. • Path Features (PF) The tree path σ(s 1 ) σ(s) connecting σ(s 1 ) and σ(s), σ(s 2 ) σ(s) conn...

Ngày tải lên: 20/02/2014, 07:20

... 2005b) and post inference (Pun- yakanok et al., 2004). These feature-based methods are considered as the state of the art method for SRL and achieved much success. However, as we know, the standard flat ... kernels, more and more kernels for restricted syntaxes or specific do- mains, such as string kernel for text categoriza- tion (Lodhi et al., 2002), tree kernel for syntacti...

Ngày tải lên: 20/02/2014, 12:20

... 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 ... be retrained and tuned for each task. On the other hand, generative models produce complete prob- ability distributions of the data, and hence can be integrated with other systems...

Ngày tải lên: 07/03/2014, 22:20

Ngày tải lên: 09/03/2014, 01:20

... or join of u, v ∈ X, if one exists, and u  v for the greatest lower bound or meet. If we need a second partial order, we use  for its order relation and  for its join operation. We are espe- cially ... encodings that use sets with intersection for meet and empty set for failure: with Y defined as the power set of some set Z,  as ⊇,  as ∩, and g(f(u), f(v)) = 0 if and o...

Ngày tải lên: 17/03/2014, 00:20

... designating the source and target languages as “French” and “English,” respectively. The variables f and e stand for source and target sentences; f j i stands for the substring of f from position ... standing for X spanning f j i . We choose b and β to balance speed and performance on our development set. For our experiments, we set b = 40, β = 10 −1 for X cells, and b...

Ngày tải lên: 17/03/2014, 05:20

... 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 learn...

Ngày tải lên: 23/03/2014, 20:21

... 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 recall is statistically significant, and it shows classifier stacking can improve performance. On th...

Ngày tải lên: 20/02/2014, 12:20

... PST as a knowl- edge source for DA detection, we essentially need to model 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 3 illustrates the PST for the sentences: Where is the Anping-Fort. There are two keywords Where and Anping-Fo...

Ngày tải lên: 20/02/2014, 05:20