Báo cáo khoa học: "Formalism-Independent Parser Evaluation with CCG and DepBank" pot

... conversion process. 249 3 The CCG Parser Clark and Curran (2004b) describes the CCG parser used for the evaluation. The grammar used by the parser is extracted from CCGbank, a CCG version of the Penn ... over-optimistic view of parser performance. In this paper we evaluate a CCG parser (Clark and Curran, 2004b) on the Briscoe and Carroll ver- sion of DepBank (Briscoe...

Ngày tải lên: 23/03/2014, 18:20

... bitexts, and one on two billion words of English. These were smoothed using modified Kneser-Ney (Chen and Goodman, 1998) and stored using randomized data structures similar to those of Talbot and Brants (2008). The ... similar to the expanded set recently used for Hiero (Chiang et al., 2009), but with bigram features (source and target word) instead of trigram features (source and...

Ngày tải lên: 07/03/2014, 22:20

... L hand to sample from and then another round with the 50 sets of randomly unsupervised seeds, S rand . The next decision is how to sample S rand from L hand . One approach is to use uniform random sampling ... and WMEB using just the hand-picked seeds (S hand ) and 50 sample super- vised bagging (S gold BAG). Bagging with samples from S gold successfully increased the performance of...

Ngày tải lên: 08/03/2014, 00:20

... each sampled sentence and its left and right phrases (see Section 3.1). We did not find sentences for 108 rules, and thus we ended up with 646 unique rules that could be evaluated (with 8945 examples ... judges. Yet, our evaluation target is to assess rules, and the Kappa values for the final correctness judgments of the shared rules were 0.74 and 0.68 for the lower and upper b...

Ngày tải lên: 08/03/2014, 02:21

... Krit1 F1 and F2 to F3 subdomains based on homology with radixin F1 in complex with IP3 and with talin F2 to F3 (Fig. 1B). These models displayed very stable secondary structures and energies with respect ... fusions in (A) were immunoblotted with GST antibodies. GST fusions in (B) and (C) were stained with Ponceau Red. Krit1-NTer and ICAP-1 were immunoblotted with His-...

Ngày tải lên: 16/03/2014, 05:20

... (Kudo and Matsumoto, 2003) for chunking, the C&C tools (Clark and Cur- ran, 2007) for CCG parsing and named entity recog- nition, and the Stanford parser (Klein and Manning, 2003a; Klein and ... the ASO model successfully finds and corrects this mistake. 8.1 Manual Evaluation We carried out a manual evaluation of the best ASO models and compared their output with...

Ngày tải lên: 17/03/2014, 00:20

... relations and thematic relations. Both relations identify related words useful to language under- standing and generation, information re- trieval, and so on. However, although words with taxonomical ... “is-a” relation (Hearst 1992), “part-of” relation (Berland and Charniak 1999), causal relation (Girju 2003), and entailment relation (Geffet and Dagan 2005). Related word...

Ngày tải lên: 17/03/2014, 04:20

... frames (expressed in functional roles, e.g. subject), and given a sentence with verbs, (auxiliary, modal, finite, non-finite, ), and chunks, ( , ), label all pairs ( ) with a grammatical role 2 . In this paper, ... chunks. 1 (e.g. a verb and a np), what is the grammatical relation between them (if there is any). We have trained a maximum entropy classifier on vectors with mor- pholog...

Ngày tải lên: 17/03/2014, 22:20

... the translation candidate is a complete sentence or just a span of it, whether the candidate is the same as or similar to the supporting candidates, and whether the supporting candidates come ... Let us start with the model in Alexandrescu and Kirchhoff (2009). According to them, a node in the graph represents the pair of some source sentence/span  and its translation candid...

Ngày tải lên: 23/03/2014, 14:20

... input to the parser is a set of system documents and the output is testcase information. The parser has two main domain- independent components, one a testing knowl- edge module and one a general ... document parser which can identify and extract testable conditions in the text. This tool would also be useful in reverse en- gineering, or taking existing artifacts of a soft-...

Ngày tải lên: 23/03/2014, 20:20