Báo cáo khoa học: "Phrase Table Training For Precision and Recall: What Makes a Good Phrase and a Good Phrase Pair?" doc

Báo cáo khoa học: "Adapting Self-training for Semantic Role Labeling" pdf

... tasks or ap-plications such as information extraction, ques-tion answering, and machine translation (Surdea-nu et al., 2003). Several CoNLL shared tasks (Carreras and Marquez, 2005; Surdeanu ... amount of natural language text available at a low cost. Semi-supervised me-thods compensate the scarcity of labeled data by utilizing an additional and much larger amount of unlabeled data ... increase in F1 is achieved by 1/10th of CoNLL training data. Therefore, for training a base classifier as high-performance as possible, while simulating the labeled data scarcity with a reasonably...

Báo cáo khoa học: "Bilingual Co-Training for Monolingual Hyponymy-Relation Acquisition" pdf

... 2IterationManually Prepared Training Data for Language 1Classifier Classifier Training TrainingEnlarged Training Data for Language 1Enlarged Training Data for Language 2Manually Prepared Training ... Data for Language 2ClassifierClassifierFurther Enlarged Training Data for Language 1Further Enlarged Training Data for Language 2Translatereliable parts ofclassification results Training Training ... sprangfrom a very simple idea: perhaps training data in a language can be enlarged without much cost if wetranslate training data in another language and addthe translation to the training data...

Báo cáo khoa học: "Semi-supervised Training for the Averaged Perceptron POS Tagger" potx

... TaggerDrahom´ıra “johanka” Spoustov´ a Jan Hajiˇc Jan Raab Miroslav SpoustaInstitute of Formal and Applied LinguisticsFaculty of Mathematics and Physics,Charles University Prague, Czech ... and “un-supervised” data for “manually annotated” and “(automat-ically annotated) plain (raw) text” data, respectively, eventhough these adjectives are meant to describe the process oflearning, ... providesingle-implementation, easy-to-maintain state-of-the-art tagging tools for as many languages as pos-sible (we are currently preparing Dutch, Slovak and several other languages).19Another area of possible...

Báo cáo khoa học: "Semi-Supervised Training for Statistical Word Alignment" docx

... California4676 Admiralty Way, Suite 1001Marina del Rey, CA 90292marcu@isi.eduAbstractWe introduce a semi-supervised approachto training for statistical machine transla-tion that alternates ... semi-supervised learning.2 Data Sets and BaselineWe conduct experiments on alignment and translation tasks using Arabic/English and French/English data sets (see Table 1 for details).Both sets have training ... compared withgold standard parallel data which was manuallyaligned using a high-recall /precision methodology(Melamed, 1998), the word-level alignments pro-duced automatically have an F-measure...

Tài liệu Báo cáo khoa học: "Re-Usable Tools for Precision Machine Translation∗" pdf

... available for system development and also serve as training data for machine learning approaches. Using thediscriminant-based Redwoods approach to tree-banking (Oepen, Flickinger, Toutanova, ... Using a trial LFG treebank for Norwegian (of less than 100 annotated sen-tences), we have adapted the tools for the currentLOGON version and are now working to train onlarger data sets and evaluate ... date. Training Material A corpus of some 50,000words of edited, running Norwegian text was gath-ered and translated by three professional transla-tors. Three quarters of the material are availablefor...

Báo cáo khoa học: "Phrase Table Training For Precision and Recall: What Makes a Good Phrase and a Good Phrase Pair?" doc

... of ACL-08: HLT, pages 81–88,Columbus, Ohio, USA, June 2008.c2008 Association for Computational Linguistics Phrase Table Training For Precision and Recall: What Makes a Good Phrase and a Good ... retrieval, precision and recall issues need to be addressed with a rightbalance for building a phrase translation table. High precision requires that identiﬁed translation candi-dates are accurate, ... approaches ofbuilding a phrase translation table and show the ﬁ-nal translation results. We measure translation per-formance by the BLEU (Papineni et al., 2002) and METEOR (Banerjee and Lavie,...

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

... translation models for phrase- based statistical machine translation (SMT) systems(Koehn et al., 2003) takes unaligned bilingual train-ing data as input, and outputs a scored table of phrase pairs. ... Joint Phrase Alignment and ExtractionGraham Neubig1,2Taro Watanabe2, Eiichiro Sumita2, Shinsuke Mori1, Tatsuya Kawahara11Graduate School of Informatics, Kyoto UniversityYoshida Honmachi, ... Och, and Daniel Marcu. 2003.Statistical phrase- based translation. In Proceedings ofthe Human Language Technology Conference (HLT-NAACL), pages 48–54.Philipp Koehn, Amittai Axelrod, Alexandra...

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

... the ACL 2007 Demo and Poster Sessions, pages 65–68,Prague, June 2007.c2007 Association for Computational LinguisticsAn Approximate Approach for Training Polynomial Kernel SVMs in Linear ... Central University National Central University Ming Chuan University Taoyuan, Taiwan Taoyuan, Taiwan Taoyuan, Taiwan bcbb@db.csie.ncu.edu.tw yang@cl.ncu.edu.tw lees@mcu.edu.tw Abstract Kernel ... Marquez, 2003), shallow parsing (Kudo and Matsumoto, 2001, 2004; Lee and Wu, 2007), named entity recognition (Isozaki and Kazawa, 2002), and parsing (Nivre et al., 2006). In particular, the use...

Báo cáo khoa học: "Combining EM Training and the MDL Principle for an Automatic Verb Classiﬁcation incorporating Selectional Preferences" pot

... collocations, and verb alternations),496 and for NLP applications in sparse data situations.In this paper, we provide an evaluation based on a language model.The remainder of the paper is organised as ... verbalternations, and collocations, and (ii) as a lexicalresource for the statistical disambiguation of parsetrees.ReferencesSteven Abney and Marc Light. 1999. Hiding a Seman-tic Class Hierarchy ... machinetranslation (Prescher et al., 2000; Koehn and Hoang,2007), document classiﬁcation (Klavans and Kan,1998), and in statistical lexical acquisition in gen-eral (Rooth et al., 1999; Merlo and Stevenson,...

Báo cáo khoa học: "Partial Matching Strategy for Phrase-based Statistical Machine Translation" pptx

... semantically similar phrases. Anothermethod is to use multi-parallel corpora (Cohn and Lapata, 2007; Utiyama and Isahara, 2007) to im-prove phrase coverage and translation quality.This paper ... incorporation by performing par-tial matching for phrase translation before decod-ing. The advantage is that the main search algorithmneed not be changed. For a source phrase f, we search partiallymatched ... 2003) adopt full matchingstrategy for phrase translation, which means that a phrase pair (f,e) can be used for translating a source phrase ¯f, only iff =¯f. Due to lack of generaliza-tion...

Xem thêm