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

Báo cáo khoa học: "Using Large Monolingual and Bilingual Corpora to Improve Coordination Disambiguation" ppt

... perfor-mance after different numbers of iterations (Fig-ure 1). We therefore set k separately for each, stop-ping around 16 iterations for the combined, 51 for the monolingual, and 57 for ... we pro-pose a way to circumvent the need for labeled data.We now describe the particular monolingual andbilingual information we use for this problem. Werefer to Table 1 for canonical examples ... information into a learned deci-sion function. The cost for this ﬂexibility is typicallythe need for labeled training data. The more featureswe use, the more labeled data we need, since for linear...

Báo cáo khoa học: "Using Lexical Dependency and Ontological Knowledge to Improve a Detailed Syntactic and Semantic Tagger of English" pot

... iteven allows a model to be trained where it wouldnot otherwise be possible to train one. For the pur-poses of our experiments, we use the top 50,000predicates for each model to form the feature ... (except for thehead child). In these experiments we used theMXPOST tagger (Ratnaparkhi, 1996) combinedwith Collins’ parser (Collins, 1996) to assign parsetrees to the corpus. The parser had ... automatically acquired on-tology made by clustering words in a large corpusof unannotated text.We have chosen to use hierarchical schemes for both the automatic and manually acquired ontolo-gies...

Báo cáo khoa học: "Using a Randomised Controlled Clinical Trial to Evaluate an NLG System" doc

... NLGSmoking Information for Heather StewartYou have good reasons to stop People stop smoking when they really want to stop. It is encouraging thatyou have many good reasons for stopping. The ... that you would be able to stop if youwere to try, you have several things in your favour.• You have stopped before for more than a month.• You have good reasons for stopping smoking.• You ... able to stop.Most people who stop smoking for good have more than one attempt.Overcoming your barriers to stopping You said in your questionnaire that you might find it difficult to stopbecause...

Báo cáo khoa học: "Using Document Level Cross-Event Inference to Improve Event Extraction" potx

... corresponds to troop movements prior to or in response to attacks. 5 Cross-event Approach In this section we present our approach to using document-level event and role information to improve ... applied to the remaining mentions in the sentence; for any argument passing that classifier, the role classifier is used to assign a role to it. Finally, once all arguments have been assigned, ... needed to understand how facts tie together. Sometimes it is difficult even for people to classify events from isolated sentences. From the sentence: (1) He left the company. it is hard to tell...

Báo cáo khoa học: "Using Syntactic Dependency as Local Context to Resolve Word Sense Ambiguity" potx

... called selectors of w: Selectorsw = ([JlceLC,~ C(Ic) ) - {w}. Step C. Select a sense s of w that maximizes the similarity between w and Selectors~. Step D. The sense s is assigned to all occurrences ... sources are used for all words, as opposed to using a separate classifier for each individual word. • It requires a much smaller corpus that needs not be sense-tagged. • It is able to deal with ... Dependency as Local Context to Resolve Word Sense Ambiguity Dekang Lin Department of Computer Science University of Manitoba Winnipeg, Manitoba, Canada R3T 2N2 lindek@cs.umanitoba.ca Abstract Most...

Tài liệu Báo cáo khoa học: "Using Smaller Constituents Rather Than Sentences in Active Learning for Japanese Dependency Parsing" docx

... vectors” for AP as well as SVM. “Support vectors”of AP means vectors which are selected in the training phaseand contribute to the prediction.8Thus it is very important to construct models for ... annotators do not give labels. Instead, labels are given virtually from correct ones that theKyoto University Corpus has.360Bunsetsu features for modiﬁers rightmost content word, rightmost function ... Mat-sumoto, 2003) for English and (Kudo and Mat-sumoto, 2002; Iwatate et al., 2008) for Japanese.Therefore, effective use of smaller constituents inactive learning would not be limited to the...

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

... Association for Computational LinguisticsUsing Machine-Learning to Assign Function Labels to Parser Output for SpanishGrzegorz Chrupała1and Josef van Genabith1,21National Center for Language ... learning to perform various graph transformations. One of thetransformations is node relabelling, which adds function tags to parser output. They report an f-score of 88.5% for the task of function ... architecture for the baseline.limited control over, or room for improvement in,this task.3.2 Adding Function Tags to Parser Output The solution we adopt instead is to add Cast3LBfunctional tags to...

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 ... andsentiment features, and therefore the expansion can-didates selected for the lexical elements in the tar-get domain reviews are sensitive to sentiment labels assigned to reviews in the source domain....

Tài liệu Báo cáo khoa học: "Using Cross-Entity Inference to Improve Event Extraction" docx

... applied to the remaining mentions in the candidate; for any argument passing that classifier, the role classifier is used to assign a role to it. Finally, once all arguments have been assigned, ... the documents to roughly describe back-ground of entity. After establishing the vector space model (VSM) for each entity mention of the type, we adopt a clustering toolkit (CLUTO) to further ... in-formation is not sufficient to make a confident de-cision. The solution is to mine credible evidences of event occurrences from global information and regard that as priori knowledge to...

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 acomparable PCFG. For example, the adaptor gram-mars for syllable structure presented in sections 3.3and 3.6 learn more information ... 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 ... 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...

Xem thêm