Báo cáo khoa học: "Underspecifying and Predicting Voice for Surface Realisation Ranking" pot

... 19-24, 2011. c 2011 Association for Computational Linguistics Underspecifying and Predicting Voice for Surface Realisation Ranking Sina Zarrieß, Aoife Cahill and Jonas Kuhn Institut f¨ur maschinelle ... methodological chal- lenge for data-driven surface realisation. The set of candidate realisations available for ranking will in- fluence the findings, and here, existin...

Ngày tải lên: 23/03/2014, 16:20

... of the dif- ferent disambiguation candidates for the same sur- face form s were very similar. For example, con- sider the disambiguation candidates of “China’ and their TF-IDF summaries in Figure ... and Witten, 2008b). In BOT, we compare the set of titles output for a doc- ument with the gold set of titles for that document (ignoring duplicates), and utilize standard precision,...

Ngày tải lên: 17/03/2014, 00:20

... structures for rela- tional learning from questions and answers. We designed sequence kernels for words and Part of Speech Tags which capture basic lexical seman- tics and basic syntactic information. ... can be used for questions and for answers, we used (and summed together) the same kernels ex- cept for those based on PASs, which are only used on answers. 4.1.1 Datasets...

Ngày tải lên: 22/02/2014, 02:20

... monomer. Monomers A (green) and B (yellow) form dimer 1, while monomers C (red) and D (blue) form dimer 2. L5 ⁄ 7, the loop between b-strands 5 and 7 which contains b-strand 6; N term, amino terminus ... dimers, with mono- mers A and B in dimer 1 and C and D in dimer 2 (Fig. 8). The a-crystallin domain of each monomer is composed of nine b-strands (labeled b2–b10), with the b6 str...

Ngày tải lên: 07/03/2014, 12:20

... original and functional BMP ⁄ activin path- way in lophotrochozoans, for which a singular and promiscuous type II receptor would be the shared interface for both BMP and activin ligands. Results Type ... pleiotropic and promis- cuous receptor. Evidence for the conservation of the mechanism of mesoderm induction in molluscs and across bilaterians Although Lartillot and collea...

Ngày tải lên: 07/03/2014, 21:20

... and short- hand were expanded, for example defib expands to defibrillator. Table 3 shows some unknown to- kens and their resolutions. The proofreading re- quire considerable amount of human effort ... ma- chine learner is that it can utilise both the infor- mation form of the concepts themselves and the contextual information, and it is able to perform prediction without seeing the en...

Ngày tải lên: 08/03/2014, 01:20

... tree. For each verb lemma, associated duration information was collected for episodic and habitual uses of the verb. Summary statis- tics for 483 verb lemmas and their typical habit and episode ... information about common events. For example, (1) and (2) were used to pro- vide information about how long a work event can last: (1) Had work for an hour and 30 mins now...

Ngày tải lên: 16/03/2014, 20:20

... one for writing programs, one for the system’s response as it executes program, and third for output tables and charts. unsupervised (+ constraints): Screen is divided into three sections: one for ... one for the system’s response as it executes program, and a third for output tables and charts. semi-supervised: The SAS screen is divided into three sections: one for writ...

Ngày tải lên: 17/03/2014, 05:20

... Generalization and Refinement Phrases are acquired in a process of hypothesis formation and error correction. The program generates and refines hy- potheses about both the linguistic pattern, and the ... meanings given for the single words carry and water are processed literally. Second, the context which exists prior to the application of the phrase, pro- vides a hypothe...

Ngày tải lên: 24/03/2014, 02:20

... schemes and parse se- lection can improve performance on biomedical data. Lease and Charniak (2005) use the Charniak parser for biomedical data and find that the use of out-of-domain trees and in-domain ... (WSJ) and unlabeled (NANC) out-of-domain data and labeled in-domain data (BROWN). Unfortunately, lacking a corresponding corpus to NANC for BROWN, we cannot perform the oppo...

Ngày tải lên: 31/03/2014, 01:20