Báo cáo khoa học: "Semantic Role Labeling Using Different Syntactic Views∗" pdf

... 581–588,Ann Arbor, June 2005.c2005 Association for Computational LinguisticsSemantic Role Labeling Using Different Syntactic Views∗Sameer Pradhan, Wayne Ward,Kadri Hacioglu, James H. MartinCenter ... investigate ways to combinehypotheses generated from semantic role taggerstrained using different syntactic views – one trained using the Charniak parser (Charniak, 2000), anotheron a rule-based ... sub-problems: the identiﬁcation of syntactic constituents that are semantic roles for a given pred-icate, and the labeling of the those constituents withthe correct semantic role. A detailed error analysis...

Báo cáo khoa học: "Semantic Role Labeling Systems for Arabic using Kernel Methods" pptx

... Ohio, USA, June 2008.c2008 Association for Computational LinguisticsSemantic Role Labeling Systems for Arabic using Kernel MethodsMona DiabCCLS, Columbia UniversityNew York, NY 10115, USAmdiab@ccls.columbia.eduAlessandro ... SRL. Syntactic Case is very important in the cases of VSOand pro-drop constructions as they indicate the syn-tactic roles of the object arguments with accusativeCase. Unless the morphology of syntactic ... expected since we are using gold standard trees.We hypothesize that the rich morphological fea-tures will help more with the role classiﬁcationtask. Therefore, we evaluate role classiﬁcation...

Báo cáo khoa học: "Semantic Role Labeling via FrameNet, VerbNet and PropBank" potx

... mapped into the Agent,Recipient, Theme and Topic theta roles.Second, we build a frequency distribution ofVN thematic roles on different syntactic positions.Based on our observation and previous ... semantic role labeling onFrameNetIn the experiments involving semantic role label-ing, we used SVMs with polynomial kernels. Weadopted the standard features developed for se-mantic role detection ... that evoke the same situations. Al-though different in conception, the FN verb clus-ters and VN verb clusters have common proper-ties1:1. Different syntactic properties between dis-tinct...

Báo cáo khoa học: "Semantic Role Labeling for Coreference Resolution" pot

... Coreference Resolution Using SRL2.1 Corpora UsedThe system was initially prototyped using theMUC-6 and MUC-7 data sets (Chinchor & Sund-heim, 2003; Chinchor, 2001), using the standardpartitioning ... available to theclassiﬁer by using SRL information.In our experiments we use the ASSERTparser (Pradhan et al., 2004), an SVM based se-mantic role tagger which uses a full syntactic analysis to automatically ... Results on MUCJSEMROLE the semantic role argument-predicate pairs of REj.For the ACE 2003 data, 11,406 of 32,502 auto-matically extracted noun phrases were tagged with2,801 different argument-predicate...

Báo cáo khoa học: "Semantic Role Labeling: Past, Present and Future" ppt

...

Báo cáo khoa học: "Grammatical Role Labeling with Integer Linear Programming" pot

... With grammatical role labeling, forexample, there is no way to explicitly expressglobal constraints that, say, the verb “to give” musthave 3 arguments of a particular grammatical role. Among the ... a grammatical role can have. Here, wehave to distinguish among our two settings. Insetting one (all case roles of all frames of a verbare collapsed into a single set of case roles), wecan’t ... Grammatical Role Labeling with Integer Linear ProgrammingManfred KlennerInstitute of Computational LinguisticsUniversity...

Tài liệu Báo cáo khoa học: "The Role of Initiative in Tutorial Dialogue" pdf

... The Role of Initiative in Tutorial DialogueMark G. Core and Johanna D. Moore and Claus ZinnSchool ... investigation of initiative, tutoringstrategy (Socratic vs. didactic), and learning (tasksuccess) using a series of human-human tutor-ing dialogues from an earlier project (Rosé et al.,2000). ... tutoring sessions for a re-search methods course. Graesser and Person de-veloped a taxonomy of different question types.Of specific interest are deep-reasoning and knowl-edge deficit questions....

Báo cáo khoa học: "Automatic Sanskrit Segmentizer Using Finite State Transducers" pdf

... possible splits with weights associ-ated with each of them. We followed two different approaches to segment a Sanskrittext using sandhi1rules extracted from aparallel corpus of manually sandhi ... rulesR1, Rk−1and the constituents c1, ckare inter-dependent since a different rule sequence will re-sult in a different constituents sequence. Also, ex-cept c1and ck, all intermediate ... 13 July 2010.c2010 Association for Computational LinguisticsAutomatic Sanskrit Segmentizer Using Finite State TransducersVipul MittalLanguage Technologies Research Center, IIIT-H,Gachibowli,...

Báo cáo khoa học: "Chinese Term Extraction Using Different Types of Relevance" potx

... proposed algorithms using different features for IT domain and legal domain, respectively. The algorithm using CD alone is the same as the TF-IDF algorithm. The algorithm using CS and CD is ... This paper presents a new term extraction ap-proach using relevance between term candi-dates calculated by a link analysis based method. Different types of relevance are used separately or ... candidates and sentences by using the link analysis approach based on the HITS algorithm to achieve better performance. In this work, a new feature on the relevance between different term candidates...

Báo cáo khoa học: "Unsupervised Sense Disambiguation Using Bilingual Probabilistic Models" pdf

... foreach sense using the conditional probabilities of thewords. As in Bengio and Kermorvant (2003), thisgenerative procedure may be captured by factor-ing the joint distribution using the conditional ... Qu´ebec,Canadabengioy@IRO.UMontreal.CAAbstractWe describe two probabilistic models for unsuper-vised word-sense disambiguation using parallel cor-pora. The ﬁrst model, which we call the Sensemodel, builds on the work of Diab ... call the Concept model, is a hier-archical model that uses a concept latent variable torelate different language speciﬁc sense labels. Weshow that both models improve performance on theword...

Xem thêm