Báo cáo khoa học: "Post-ordering by Parsing for Japanese-English Statistical Machine Translation" potx

... Tree-based statistical machine translation models have made significant progress in re- cent years, especially when replacing 1-best trees with packed forests. However, as the parsing accuracy ... AFNLP Sub-Sentence Division for Tree-Based Machine Translation Hao Xiong * , Wenwen Xu + , Haitao Mi * , Yang Liu * and Qun Liu * * Key Lab. of Intelligent Information Processing...

Ngày tải lên: 31/03/2014, 00:20

... criteria are, for example, F-Measure for parsing, mean average precision for ranked retrieval, and BLEU or multi-reference word error rate for statistical machine translation. The use of statistical ... been applied for exam- ple in parsing (Goodman, 1996) and word alignment (Kumar and Byrne, 2002). 9 Conclusions We presented alternative training criteria for log- linear...

Ngày tải lên: 23/03/2014, 19:20

... language model for speech recognition was measured both by parsing accuracy (parsing strings with annotated reference parses), and by WER. WER is measured by parsing word lattices and comparing the sentence yield ... model parameters. This score, defined solely by the parsing model, will be referred to as the parser score. The total score for chart edges for the lattice p...

Ngày tải lên: 20/02/2014, 15:21

... step, thus leading to inferior performance. Bottom-up parsing is far more attractive for lexi- calist formalisms, as it is driven by the syntactic in- formation associated with lexical elements. ... argues that head-corner parsing is especially useful for parsing with non-concatenative grammar formalisms. In [Lavelli and Satta, 1991] a head-driven parsing strategy for L...

Ngày tải lên: 09/03/2014, 01:20

... July 2006. c 2006 Association for Computational Linguistics Clavius: Bi-Directional Parsing for Generic Multimodal Interaction Frank Rudzicz Centre for Intelligent Machines McGill University Montr ´ eal, ... expanded by Γ 2 : NP −→ DT NN. 1.2 Multimodal Bi-Directional Parsing Our parsing strategy combines bottom-up and top-down approaches, but differs from other approaches to b...

Ngày tải lên: 23/03/2014, 18:20

... results for the immediate- bihead model 4. all words except for the 10,000 most com- mon are replaced by the symbol UNK. As in previous work, files F0 to F20 are used for training, F21-F22 for development, ... ones for which the trigram model performs better, and the ones for which the grammar model does better. We then collect labeled precision and recall statistics (the stan- d...

Ngày tải lên: 31/03/2014, 04:20

... partly supported by SFB 314, Project N3 BiLD; and by the NBBI via the Eu- rotra project. I am grateful to Mike Reape for useful com- ments, and an anonymous reviewer of ACL, for pointing out ... and the CKY algorithm, for the same class of grammars. I present a head-driven bottom-up algorithm for F-LCFR grammars. The algorithm resembles the head-driven parser by Martin K...

Ngày tải lên: 31/03/2014, 06:20

... the most representative work is the forest-based translation method (Mi et al. 2008; Mi and Huang 2008; Zhang et al. 2009) in which a packed forest (forest for short) structure is used to effectively ... mentioned before, tree-to-string approaches are more vulnerable to parsing errors. In practice, the Berkeley parser (Petrov et al. 2006) we used yields unsatisfactory parsing p...

Ngày tải lên: 17/03/2014, 00:20

... language. We obtain substantial improvements in performance for translation from Chinese and Arabic to English. 1 Introduction The statistical revolution in machine translation, be- ginning with (Brown ... Since component features in those combi- nations were informed by individual-feature perfor- mance on the test set, we tested the best perform- ing conditions from MT06 on a new tes...

Ngày tải lên: 17/03/2014, 02:20

... dialogue act tag set, which is then used for train- ing the statistical models for automatic dialogue act classification. 3.1 Tag Set We chose 12 tags by manually labelling the dialogue corpus ... represented by NO- ANSWER in our tag set. The Kappa statistic was used to compare inter- annotator agreement normalised for chance (Siegel and Castellan, 1988). Labelling was carried out...

Ngày tải lên: 17/03/2014, 06:20