Báo cáo khoa học: "Using comparable corpora to solve problems difﬁcult for human translators" pptx

... 2006.c2006 Association for Computational LinguisticsUsing comparable corpora to solve problems difﬁcult for human translatorsSerge Sharoff, Bogdan Babych, Anthony HartleyCentre for Translation StudiesUniversity ... present a tool that uses comparable corpora to ﬁnd appropriatetranslation equivalents for expressions thatare considered by translators as difﬁcult. For a phrase in the source language thetool ... translations for polysemouswords are too numerous to be listed for all pos-sible contexts. For example, the entry for strongin ORD already has 57 subentries and yet it fails to mention many...

Báo cáo khoa học: "Using Similarity Scoring To Improve the Bilingual Dictionary for Word Alignment" doc

... 50performanceminscore’Precision1-Dict0.01-Cog7.3’’Precision1-Dict0.01-Baseline’’Recall1-Dict0.01-Cog7.3’’Recall1-Dict0.01-Baseline’’F-measure1-Dict0.01-Cog7.3’’F-measure1-Dict0.01-Baseline’Figure 4: Performance of dictionaries Dict0.01 for up to one link per wordrebuilding algorithm is independent of the actualword alignment method used.Furthermore, we plan to explore ways to improvethe ... is to enable variations of a word to receivea higher dictionary score than words that are verycommon overall.Furthermore, we have decided to exclude wordsfrom clustering that account for ... cooccurrencescore with has to be in relation to all other scoresbetween and a target language word.is expressed as follows: a word qualiﬁes for clus-tering ifAs before, are all the target language...

Báo cáo khoa học: "Using bilingual dependencies to align words in Enlish/French parallel corpora" ppt

... are generally fusiform. Le corps des aloses est généralement fusiforme. (4) Ability to react to light. Capacité à réagir à la lumière. UNAMBIGUOUS GOV -TO- DEP PROPAGATION TO ALIGN NOUNS. ... and fall to the bottom. Les oeufs de très petite taille tombent sur le fond. (5) X is a model which was designed to stimulate… X est un modèle qui a été conçu pour stimuler… GOV -TO- DEP ... propagated to their dependents (Gov -to- Dep propagation). In this case, several relations that may be used to achieve the propagation are avail-able, as it is possible for a governor to have more...

Tài liệu Báo cáo khoa học: "Using Multiple Sources to Construct a Sentiment Sensitive Thesaurus for Cross-Domain Sentiment Classiﬁcation" doc

... agnostic to theproperties of the classiﬁer and can be used to expandfeature vectors for any binary classiﬁer). L1 regular-ization enables us to select a small subset of features for the classiﬁer. ... automatic method to createa thesaurus that is sensitive to the sentiment ofwords expressed in different domains.• We describe a method to use the created the-saurus to expand feature vectors ... vector d ∈ RN, where thevalue of the j-th element djis set to the total numberof occurrences of the unigram or bigram wjin thereview d. To ﬁnd the suitable candidates to expand avector...

Tài liệu Báo cáo khoa học: "Using adaptor grammars to identify synergies in the unsupervised acquisition of linguistic structure" docx

... structure needs to be “built in” to an adaptor grammar compared to a comparable PCFG. For example, the adaptor gram-mars for syllable structure presented in sections 3.3and 3.6 learn more information ... never expands to itself); it iscurrently unknown whether there are tree distribu-tions that satisfy the adaptor grammar constraints for recursive adaptor grammars.Inference for an adaptor grammar ... con-textual information is less important for their acqui-sition than, say, syntax.2 From PCFGs to Adaptor GrammarsThis section introduces adaptor grammars as an ex-tension of PCFGs; for a more...

Tài liệu Báo cáo khoa học: "Using linguistic principles to recover empty categories" ppt

... of rules. 1 for each tree, iterate over nodes from top down 2 for each node X 3 try to insert NP* in X 4 try to insert 0 in X 5 try to insert WHNP 0 or WHADVP 0 in X 6 try to insert *U* ... try to insert a VP ellipsis site in X 8 try to insert S*T* or SBAR in X 9 try to insert trace of topicalized XP in X 10 try to insert trace of extraposition in X 11 for each node X 12 try to ... if the task is to insert empty nodes into a tree, then the method leads both to false positives and to false negatives. Suppose for example that the sentence When do you expect to finish? has...

Tài liệu Báo cáo khoa học: "USING BRACKETED PARSES TO EVALUATE A GRAMMAR CHECKING APPLICATION" ppt

... developed a methodology for using hand-bracketed parses to examine both the inter- nal and external performance of a grammar checker. The internal performance refers to the behavior of the ... we therefore weight repaired to be preferred as a passive verb. Although this method for selecting fronted parse trees some- times leads to false error critiques, it works well for most ... In order to coerce our system into accepting only the desired parse tree, we modified it to ac- cept only parses that satisfied bracketed forms. 6. The BSEC has the capability to report on...

Báo cáo khoa học: "Using Machine-Learning to Assign Function Labels to Parser Output for Spanish" ppt

... function tag and trace informa-tion. In contrast to English, in many other lan-guages conﬁgurational information is not a goodpredictor for LFG grammatical function assign-ment. For such languages ... memory-based learning to perform various graph transformations. One of thetransformations is node relabelling, which addsfunction tags to parser output. They report an f-score of 88.5% for the task ... Parsing For constituency parsing we use Bikel’s (2002)parser for which we developed a Spanish languagepackage adapted to the Cast3LB data. Prior to parsing, we perform one of the tree transforma-tions...

Báo cáo khoa học: "Using Grammatical Relations to Compare Parsers" pptx

... function tags to PennTreebank I style parses; however, as well asconverting the tags into a Carroll et al. for- mat, we would need to add extra rules to ex-tract other GRs needed for our application ... use GRs to provide a common basis for comparing full and shallow parsers, andPenn Treebank and Susanne structures. To carry out this comparison, we implemented aGR extraction mechanism for Penn ... of 500 Susanne sentences due to Carroll et al. To gain insight into the strengthsand weaknesses of the different parsers, wepresent a breakdown of the results for eachtype of GR. It is not...

Báo cáo khoa học: "Using Search-Logs to Improve Query Tagging" potx

... sentences andalso the automatically generated query training set.The simplest way to match query tokens to snip-pet tokens is to allow a query token to match anysnippet token. This can be problematic ... strate-gies for selecting which annotation to transfer andﬁnd that using the result that was clicked by the usergives comparable performance to using just the topresult or to aggregating over the top-k ... has the ability to pool knowledge from several related queries andincorporate the information into the model param-eters. An area for future work is to transfer othersyntactic information, such...

Xem thêm