Báo cáo khoa học: "Identifying Broken Plurals, Irregular Gender, and Rationality in Arabic Text" ppt
... that training single classifiers and combining their outcomes almost always outperforms a single joint classi- fier for the three target features. In other words, combining the results of G and N ... common irregular word forms, such as broken plurals (which resemble singular nouns), and nouns with irregular gender (feminine nouns that look masculine and vice versa). In additi...
Ngày tải lên: 24/03/2014, 03:20
... 1 after deducting basal and background values. Sepsis resulted in a significant increase in the transcription of the mRNA encoding iNOS in heart of iNOS + ⁄ + mice. The expression of iNOS mRNA was ... damaged during the early and acute phases of Chagasic cardiomy- opathy [41]. In these phases, the innate in ammatory response corresponds with iNOS induction and a subse- quent inc...
Ngày tải lên: 19/02/2014, 00:20
... efficient, language pair independent and mines transliteration pairs in a consistent fashion in both unsupervised and semi-supervised settings. We model transliter- ation mining as an interpolation of ... the unlabelled data to help correctly align it and train our unsupervised min- ing system on the combined labelled and unlabelled training data. In the expectation step, the pri...
Ngày tải lên: 19/02/2014, 19:20
Tài liệu Báo cáo khoa học: "Creating a manually error-tagged and shallow-parsed learner corpus" pptx
... major cause of false pos- itives and negatives in error detection may be at- tributed to errors in POS-tagging and chunking. In corpus linguistics, researchers (Aarts and Granger, 1998; Granger, 1998; ... existing POS-tagging/chunking tech- niques on learner data. We report and discuss the results in Sect. 5. 2 Difficulties in Learner Corpus Creation In addition to the comm...
Ngày tải lên: 20/02/2014, 04:20
Tài liệu Báo cáo khoa học: "SenseLearner: Word Sense Disambiguation for All Words in Unrestricted Text" doc
... For instance, using the SENSELEARNER learn- ing mechanism, a model can be defined and trained to handle all the nouns in the test corpus. Similarly, us- ing the same mechanism, a finer-grained ... model-specific, and feature vectors are added to the training set pertaining to the corresponding model. The label of each such feature vector consists of the target word and the correspond- in...
Ngày tải lên: 20/02/2014, 15:20
Tài liệu Báo cáo khoa học: "Linear Context-Free Rewriting Systems and Deterministic Tree-Walking Transducers*" pptx
... deterministic tree-walking transducers have been well studied (see [4]). In the remainder of the paper we describe linear context-free rewriting systems and deterministic tree- walking transducers ... semilinear and can be recognized in polynomial time. The definition of lin- ear context-free rewriting systems is deliberately not specific about the kinds of structures being man...
Ngày tải lên: 20/02/2014, 21:20
Báo cáo khoa học: "A Corpus for Modeling Morpho-Syntactic Agreement in Arabic: Gender, Number and Rationality" docx
... two broken plu- rals: ktAb ( M S M P ) 4 and ktb¯h ( F S M P ). In ad- dition to broken plurals, Arabic has a class of bro- ken feminines in which the feminine singular ... nouns and adjectives) and verbs in ect for gender: mascu- line (M) and feminine (F ), and for number: sin- gular (S), dual (D) and plural (P ). These fea- tures are typical...
Ngày tải lên: 07/03/2014, 22:20
Báo cáo khoa học: "Multimodal Menu-based Dialogue with Speech Cursor in DICO II+" pptx
... which in turn is implemented using TrindiKit (Traum and Larsson, 2003). The main goal of the original Dico application (Olsson and Villing, 2005), (Villing and Larsson, 2006) was to develop an interface ... dis- tracting for the driver, and thus both safer and easier to use than existing interfaces. (Larsson and Villing, 2009) described the Dico II system resulting from work in...
Ngày tải lên: 07/03/2014, 22:20
Báo cáo khoa học: "Generating Usable Formats for Metadata and Annotations in a Large Meeting Corpus" pptx
... of relevant information, which is done here by solving NXT pointers and discarding NXT-specific markup to group all information for a phenomenon in only one structure or table. The following criteria ... The stylesheets resolve most of the NXT pointers, by including redundant information into the tables, in order to speed up queries by avoiding frequent joins. A Perl script applies the r...
Ngày tải lên: 08/03/2014, 02:21
Báo cáo khoa học: "Automatically Creating Bilingual Lexicons for Machine Translation from Bilingual Text" ppt
... Computational Linguistics, pages 449-451, Helsinki, Finland. D. Turcato, O. Laurens, P. McFetridge, and F. Popowich. 1997. Inflectional information in transfer for lexicalist MT. In Proceed- ings ... The information con- tained therein is used to assign preferences to competing candidate entries, in two ways. Firstly, templates are probabilistically ranked, using the existin...
Ngày tải lên: 08/03/2014, 06:20