Báo cáo khoa học: "A Meta-Level Grammar: Redefining Synchronous TAG for Translation and Paraphrase" doc

... A Meta-Level Grammar: Redefining Synchronous TAG for Translation and Paraphrase Mark Dras Microsoft Research Institute Department ... how a meta-level gram- mar resolves the representational problems; and it further shows that this has no effect on the generative capacity of S -TAG. 2 S -TAG and Machine Translation Synchronous ... object-level formal- ism wo...

Ngày tải lên: 08/03/2014, 06:20

... corpus that were not used for training. The POS tagger we developed is currently the fastest tagger for the Portuguese language, and it is in line with state-of-the-art taggers for other lan- guages, ... a contraction. 5 4 POS tagger For the POS tagging task we used Brant’s TnT tag- ger (Brants, 2000), a very efficient statistical tag- ger based on Hidden Markov Models. For tra...

Ngày tải lên: 17/03/2014, 22:20

... specifically, in terms of c-command and minimal dominating S and NP domains, control and unbounded de- p endency structures can only be handled b~' ad- itional and fairly complex devices. ... Reinhart (1976) and (1983) for alternative definitions of c-command, and discussions of the role of this re- lation in determining the possibilities of anaphora. See Lappin (1985)...

Ngày tải lên: 17/03/2014, 20:20

... language processing (Lari and Young, 1990; Pereira and Schabes, 1992; Carroll and Charniak, 1992). The constituent con- text model (CCM) for inducing constituency parses (Klein and Manning, 2002) was ... dependency grammar induction. For instance, the dependency model with valence (DMV) of Klein and Manning (2004) has been extended to utilize multilingual in- formation (Berg-Kir...

Ngày tải lên: 23/03/2014, 14:20

... of endings: andare (to go), for example, admits two different stems, vad (go) and and (went); so there exist two subclasses of endings named respectively dvl andare and dr2 andare. 4. ... di di prep.simple 2 andare vad dv 1 _andare v.intran.simple 1 andare and I dv2. andar© v.intran.simple I The other sets of data are contained in the Prolog workspace and are structur...

Ngày tải lên: 24/03/2014, 05:21

... divided by the sum of intensities for full-length and truncated ApoC1. The rate constant for disappearance of ApoC1 was determined from a first-order decay plot, and standard deviation from triplicate ... dithio- threitol and iodoacetamide, with quantification of Cys by detection of mass change of a peptide after reduc- tion and alkylation. These and other approaches have been des...

Ngày tải lên: 19/02/2014, 05:20

... Juro Supplies, Lucerne, Switzerland) were added, and the slurry was incu- bated for 36 h at 6 °C on a rotating wheel. The beads were then pelleted by centrifugation, and the supernatant con- taining ... kDa) and cata- lyze (232 kDa). Acknowledgements We would like to thank M. Linder and X. Lan Vu for their technical assistance, R. Brun of the Swiss Trop- ical Institute, University...

Ngày tải lên: 19/02/2014, 07:20

... non-copula questions and build the model for only copula questions. ponent of a candidate sentence. For example for the given question, ”When did Nixon die?”, when the following candidate sentence, ... affirmed ques- tions did not contain any object and they are also in copula (linking) sentence form that is, they are only formed by subject and information about the subject as: {su...

Ngày tải lên: 20/02/2014, 07:20

... in the form of “Mr. and Mrs. Smith”, “Mr. And Mrs. Smith”, and so on. To deal with these aliasing forms, we transform all different forms into a standard form for the later ranking and identification. ... retrieve candi- dates and then use lexical information, frequen- cies, and distances to find the correct translation. They achieve a Top-1 accuracy of 81.0% and claim t...

Ngày tải lên: 20/02/2014, 12:20

... instances for the start classifier, 500,000 for the end classifier, and 272,000 for the role classifier. The features used by the classifiers are described in Subsec- tion 3.2, and the performance ... A list L of chunks and a target word t Binary classifiers starts and ends Output: The sets S and E of start and end brackets Split L into the sublists L before , L target , and L...

Ngày tải lên: 20/02/2014, 12:20