Báo cáo khoa học: "An Online System for Corpus Management and Analysis in Support of Computing in the Humanities" pot

... aspect of pure resource management, processing and analysis of docu-ments have traditionally been the domain of desk-top applications. Sometimes even to the point of command line tools. Therefore ... Frankfurt am Main,2Universit¨at BielefeldAbstractThis paper introduces eHumanities Desk-top- an online system for corpus manage-ment and analysis in support of Comput-ing in the Humanities. ... store the resource.Since the output of the tagger is a XML documentit is stored as a XML Database. Finally the in- formation about the new document is stored in the Master Data including a...

Báo cáo khoa học: "An Unsupervised System for Identifying English Inclusions in German Text" doc

... the classiﬁ-cation results of our system and comparethem to the performance of a trained ma-chine learner in a series of in- and cross-domain experiments.1 Introduction The recognition of ... used and the architecture of our system. In Sec-tion 5, we provide an evaluation of the system out-put and compare the results with those of a series of in- and cross-domain machine learning ... While the tagger performs much better for the internet and the space travel data, it requires hand-annotatedtraining data. The lookup system, on the other hand,is essentially unsupervised and therefore...

Báo cáo khoa học: "AN EXPERT SYSTEM FOR THE PRODUCTION OF PHONEME STRINGS FROM UNMARKED ENGLISH TEXT USING MACHINE-INDUCED RULES" pdf

... consists of all of the entries from the SCL and the SEL, as well as all of the features for each entry. At this point an initial "window" on the training set is chosen. Since the inference ... t~ next word in the input. ~. Traininm Mode When UTTER is operating in training mode, the system allows the user to correct errors in transcription interactively by specifying the proper ... execution time increases comblnatorlally with the size of the training set, it is wise to begin the inference procedure with a subset of the training set. This is acceptable since there is often a...

Tài liệu Báo cáo khoa học: "An Unsupervised Model for Joint Phrase Alignment and Extraction" ppt

... ComputationalLinguistics, pages 25–28.John DeNero and Dan Klein. 2010. Discriminative mod-eling of extraction setsfor machine translation. In Pro-ceedings of the 48th AnnualMeeting of the Association for ... Ph.D. thesis, Massachusetts Institute of Tech-nology.John DeNero and Dan Klein. 2008. The complexity of phrase alignment problems. In Proceedings of the 46thAnnual Meeting of the Association for ... model. For GIZA ++, we use the standard training reg-imen up to Model 4, and combine alignmentswith grow-diag-final -and. For the proposedmodels, we train for 100 iterations, and use the ﬁnalsample...

Tài liệu Báo cáo khoa học: "An Ensemble Method for Selection of High Quality Parses" pdf

... both the generative parsingmodel number 2 of Collins (1999) and the rerankingparser of Charniak and Johnson (2005), both when the training and test data belong to the same domain (the in- domain ... the way f-score is ordinarily calculated, by computing the labeled precision and recall of the constituents in the whole set and using these as the arguments of the f-score equation. The ordinary ... copies of a parser, each of which trained on a different sample from the trainingdata, to predict the quality of a parse. The algorithmdoes not assume uniformity of training and test data,and...

Tài liệu Báo cáo khoa học: "An Approximate Approach for Training Polynomial Kernel SVMs in Linear Time" doc

... vector in D-dimension space of the i-th example, and yi is the label of xi either positive or negative. The training of SVMs involves in minimize the following object (primal form, soft-margin) ... trade off training error and mar-gin. A small value for C will increase the number of training errors. To determine the class (+1 or -1) of an example x can be judged by computing the following ... support in the training set and lim-ited in the length of predefined d, which is equiva-lent to the polynomial kernel degree d. For exam-66ple, if the minimum support is 5, and d=2, then a...

Tài liệu Báo cáo khoa học: "An Improved Parser for Data-Oriented Lexical-Functional Analysis" doc

... respect to the corpus& quot;, thus increasing the robustness of the model.2.3 The composition operation In LFG-DOP the operation for combiningfragments is carried out in two steps. First the c-structures ... using initially the indexed subtrees only.Thus only the Category-matchingcondition is enforced during the chart-parsingprocess. The Uniqueness and Coherenceconditions of the corresponding ... randomexcept for one constraint: that all the words in the test set actually occurred in the training set. The sentences from the test set were parsed and disambiguated by means of the fragments fromthe...

Tài liệu Báo cáo khoa học: "An Integrated Environment for Computational Linguistics Experimentation" pot

... for theoretical models, and the accuracy of these models can be evalu-ated either with regard to their ability to account for the reality of a given corpus (pursuing descrip-tive aims), either ... models and methodological concerns. Finally,when other platforms usually enforce the use of adedicated document format, LinguaStream is ableto process any XML document. On the other hand,LinguaStream ... stream, the platform takes care of the scheduling of sub-tasks, and various tools allow the results to be visualisedconveniently.Fundamental principlesFirst of all, the platform makes use of declarativerepresentations,...

Tài liệu Báo cáo khoa học: "An expressive formalism for describing tree-based grammars" docx

... con-sists of tree description and/ or of semantic formu-las. The XMG formalism furthermore supports the sharing of identiﬁers across dimension hence al-lowing for a straightforward encoding of the ... semantics The XMG formalismfurther supports the integration in the grammar of semantic information. More generally, the lan-guage manages dimensions of descriptions so that the content of a class ... factorizing grammar in- formation is crucial for the rapid development, the maintenance and the debugging of large scalegrammars. While much work has been deployedinto producing such factorizing...

Tài liệu Báo cáo khoa học: "An annotation scheme for discourse-level argumentation in research articles" doc

... amount of training, including the reading of coding instructions for the two versions of the scheme (6 pages for the basic scheme and 17 pages for the full scheme), four training papers and ... determining BASIS and CONTRAST. This might have to do with the loca- tion of those types of sentences in the paper: AIM and TEXTUAL are usually found at the beginning or end of the introduction ... no training in computational linguistics performed reasonably well: they were not part of the circle of the three most similar subjects in their groups, but they were also not performing worse...

Xem thêm