Báo cáo khoa học: "Learning a Compositional Semantic Parser using an Existing Syntactic Parser" pot

... for learning semantic parsers do not utilize an existing syntactic parser that pro- vides disambiguated parse trees. 1 However, ac- curate syntactic parsers are available for many 1 Ge and Mooney ... semantic analyzers. This paper presents an approach to learning semantic parsers that uses parse trees from an existing s yntactic analyzer to drive the interpretation proce...

Ngày tải lên: 23/03/2014, 16:21

... Representations and Lexical Semantic Resources: Towards Accurate Deep Semantic Parsing Sergio Roa†‡, Valia Kordoni† and Yi Zhang† Dept. of Computational Linguistics, Saarland University, Germany† German ... predicate-argument structures or label semantic roles that reveal the par- tial meaning of sentences (Carreras and M ` arquez, 2005). Manually annotated lexical semantic re- source...

Ngày tải lên: 08/03/2014, 01:20

... Branavan Harr Chen Jacob Eisenstein Regina Barzilay Computer Science and Artificial Intelligence Laboratory Massachusetts Institute of Technology {branavan, harr, jacobe, regina} @csail.mit.edu Abstract This ... a wide margin in all evaluations. The absolute performance of the automatic meth- ods indicates the difficulty of the task. For instance, evaluation against gold standard annotations...

Ngày tải lên: 23/03/2014, 17:20

... replaced by an abasic analog [27]. It was observed that 2,6-diaminopurine was significantly more efficient than adenine in restor- ing the catalytic activity. In an attempt to obtain addi- tional ... [8] and lower eukaryotes, amphibians, prokaryotes, bacterio- phages, and viroids and satellite viruses that infect plants. A ribozyme also exists in hepatitis delta virus, a serious human p...

Ngày tải lên: 07/03/2014, 00:20

... sec- ond dataset contains three annotated presidential de- bates (Boydstun et al., 2011) between Barack Obama and John McCain and a vice presidential debate be- tween Joe Biden and Sarah Palin. Each ... moderator. 7 Similarly, the “Question” speaker had a relatively high variance, consistent with an amalgamation of many distinct speakers. These topic shift tendencies suggest that all ca...

Ngày tải lên: 07/03/2014, 18:20

... ihbaht:1, baot:1 “wanna” waanah:94, waanih:37, wahnah:16, waan:13, wahneh:8, wahnih:5, wahney:3, waanlih:3, wehnih:2, waaneh:2, waonih:2, waaah:1, wuhnih:1, wahn:1, waantah:1, waanaa:1, wowiy:1, waaih:1, wah:1, ... weights. “about” ahbawt:15, bawt:9, ihbawt:4, ahbawd:4, ih- bawd:4, ahbaat:2, baw:1, ahbaht:1, erbawd:1, bawd:1, ahbaad:1, ahpaat:1, bah:1, baht:1, ah:1, ahbahd:1, ehbaat:1, ahbaed...

Ngày tải lên: 16/03/2014, 19:20

... on the Charniak Parser (Charniak-Charniak). In gold standard parse trees the syntactic cate- gories are annotated with functional tags. Since cur- rent statistical parsers do not annotate sentences ... gold standard parse trees (Gold- Gold), and text annotated using a state-of-the-art sta- tistical parser (Charniak and Johnson, 2005) (Gold- Charniak), respectively. In the third, we...

Ngày tải lên: 17/03/2014, 02:20

... COMMUNICATOR data (automatically annotated with speech acts and Information States (Georgila et al., 2005b)), they then output a user utterance as a further speech-act task pair. These user simulations ... Then the baseline strategy’s state has not changed and so it will repeat its last dialogue move, whereas the state for strategies 2 (UDM) and 3 (USDM) has changed and as a result, the...

Ngày tải lên: 17/03/2014, 04:20

... relations: ⎧ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎨ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎩ RESTAURANT has foodquality RESTAURANT has servicequality RESTAURANT has valuequality RESTAURANT has atmospherequality RESTAURANT has overallquality RESTAURANT ... Linguistic Variation We also wish to assess whether the linguistic variation of the learned mappings was greater than what we could easily have generated with a hand-crafted diction...

Ngày tải lên: 23/03/2014, 18:20

... in a real-life situation. 4.1 Progressive validation When learning from data streams the standard evaluation methodology where data is split into a separate training and test set is not applicable. ... of the top ranking suggesting, but rather the quality of the whole ranked list. Previous research (Bhattacharya and Neamtiu 2010, Tamrawi et al. 2011) made an attempt at approximating thi...

Ngày tải lên: 24/03/2014, 03:20