Báo cáo khoa học: "ADP based Search Using Monotone Alignments in Statistical Translation" pdf

... Ney, and A. Zubia- ga. 1997b. A DP -Based Search using Monotone Alignments in Statistical Translation. In Proceed- ings of the ACL/EACL '97, Madrid, Spain, pages 289-296, July. S. Vogel, ... of the model. 1.3 Search in Statistical Machine Translation In the last few years, there has been a number of papers considering the problem of finding an effi- cient...

Ngày tải lên: 20/02/2014, 18:20

... given their online-nature. We also see that parsing is much faster. SVM CW Training 75 min 8 min Parsing 29 min 1.5 min Table 1: Training and parsing time on Danish data. 5.3 Pruning features Because ... ob- tained using margin -based machine learning ap- proaches. For the MST-parsing MIRA (McDonald et al., 2005a; McDonald and Pereira, 2006) and for transition -based parsing Support-Vecto...

Ngày tải lên: 20/02/2014, 04:20

... words (λ), is indeed an outside factor which confines the searching scope to the cost of computation and depends on the different applications. If we tuned it using the training data set of ... context setting into the computational one to narrow down the specific meaning of the target, manually realiz- ing this with the help of Roget’s Thesaurus. They defined a lexical chain within...

Ngày tải lên: 08/03/2014, 02:21

... associated with the words in each window are added up, and the cosine between the resulting win- dow vectors is assigned to the gap between the win- dows in the text. High values indicate lexical close- ... Considering the additional effort and resources involved in using context vec- tors, the modest boost in performance might often not be worth the effort in practice. This...

Ngày tải lên: 17/03/2014, 07:20

... activities (quest- ing, competing, fighting, etc.) and communicate via chat or speech using assumed screen names. We collected a corpus of 48 group quests in Second Life that jointly involved 206 ... work in progress. 4.3.3 Zone coding Each of the five main areas had a correspond- ing set of subzones. A subzone is a building, a room within a building, or any other identifiable ar...

Ngày tải lên: 20/02/2014, 04:20

... com- bining discriminative training with features to detect these two different kinds of move- ment phenomena leads to substantial im- provements in word ordering performance over strong baselines. ... strong baselines. Integrating this word order model in a baseline MT system results in a 2.4 points improvement in BLEU for English to Japanese translation. 1 Introduction The machine t...

Ngày tải lên: 20/02/2014, 12:20

... below, indicating the starting point and ending point of each phrase in the input string. lnput~ Sentenc(l: 0 They t are 2 flying 3 planes 4 Gram.mar: N " * they V * are N * tl¥ing ... of/d/before/y /in didyou (5b) Reduction of unstressed/u/to schwa in) ,~u (5c) Flapping of intervocalic /t/ in hit. it (5d) Reduction of schwa and devoicing of/u /in to (5e) Reduc:io...

Ngày tải lên: 21/02/2014, 20:20

... Networks" in Artificial Intelligence 13 (1980). [11] David Scott Warren, " ;Using A-Calculus to Represent Meaning in Logic Grammars" in the Proceedings of the 21st Annual Meeting of the ... genuine problem with representing the quantification. We need to represent the variable being quantified as a gen- uine variable, since, for instance, instantiating the quanti-...

Ngày tải lên: 21/02/2014, 20:20

... regarding certain aspects of automat- ically translating Russian predicative infinitives into English. Emphasis is placed on the analysis (decoding) of the pertinent infinitive constructions in ... word order in Russian. Isačenko treats some of the problems of am- biguity discussed here (dative + infinitive, infinitive with double dative) using a combination of methods including t...

Ngày tải lên: 07/03/2014, 18:20

... will incorporate it in an indi- rect way: we train a classifier C on the additional training data T  , which we then use to classify the original training data T . Then we train the all- in- one ... the all -in- one and joint models, in order to maintain fairness in com- paring the models, we modified the procedures to allow for the inclusion. Let T denote the original training data, an...

Ngày tải lên: 08/03/2014, 02:21