Báo cáo khoa học: "Learning Stochastic OT Grammars: A Bayesian approach using Data Augmentation and Gibbs Sampling" pptx

... 43rd Annual Meeting of the ACL, pages 346–353,Ann Arbor, June 2005.c2005 Association for Computational LinguisticsLearning Stochastic OT Grammars: A Bayesian approach using Data Augmentation ... a non -Bayesian perspective, the MCMC-based approach can be seen as a randomized strategy forlearning a grammar. Computing resources make itpossible to explore the entire space of grammars and discover ... function has a complex form, thus making a maximum-likelihood approach hard to perform.Such problems are avoided by using the Data Aug-mentation algorithm (Tanner and Wong, 1987) tomake computation...

Báo cáo khoa học: "Learning Semantic Links from a Corpus of Parallel Temporal and Causal Relations" doc

... These works annotated temporal relationsbetween events and times, but low inter-annotatoragreement made many TimeBank and TempEvaltasks difficult (Boguraev and Ando, 2005; Verha-gen et al., 2007). ... should label returned-waitingwith BEFORE since returned occurred first, and a causal classifier should label it CAUSAL since this and can be paraphrased as and as a result.We identified both syntactic ... semantic features.All+Tmp (Causals Only) Syntactic and semanticfeatures, plus the gold-standard temporal label.We compared our models against several baselines, using precision, recall and F-measure...

Báo cáo khoa học: "Learning the Countability of English Nouns from Corpus Data" ppt

... countability prefer-ences of English nouns from unannotated corpora.We first annotate them automatically, and then trainclassifiers using a set of gold standard data, takenfrom COMLEX (Grishman ... determine NP and PP boundaries, and medium-recall chunk adjacency templates to recover inter-phrasal dependency. Third, we fully parse the data and simply read off all necessary data from the ... thegold-standard training data from unambiguouslynegative exemplars and potentially ambiguous pos-itive exemplars. That is, we would like classifiersto judge a target noun as not belonging to a givencountability...

Báo cáo khoa học: "Learning How to Conjugate the Romanian Verb. Rules for Regular and Partially Irregular Verbs" docx

... but not all and/ or not for the same person and number.1. no alternation; a spera” (to hope);2. alternation:˘ a e for the 2nd person singular; a num˘ara” (to count);3. no alternation; a ... ”d”;18. alternation:˘ a a for all forms except the 1st and 2nd person plural, d→z for the 2nd per-son singular due to palatalization; a c˘adea”(to fall);19. no alternation; a veghea” (to watch ... verbe romˆanes¸ti gru-pate pe clase de conjugare. Bucharest: Coresi,2007. 4th edition, revised. (In Romanian.) (263pp.).Ana-Maria Barbu. Romanian lexical databases:Inﬂected and syllabic forms...

Báo cáo khoa học: "LEARNING TRANSLATION SKILLS WITH A KNOWLEDGE-BASED FRENCH-ITALIAN CONJUNCTIONS IN CONTEXT" pdf

... of ELISA's research within a project of automatic translation, and for a better understanding and explanation of the student's misconceptions as well. Because the " ;a posteriori" ... education- al settings and in projects of natural language translation. Practically, our program is one of the few Intelligent Systems available in the field of Fo- reign Language Teaching and ... well as the incorrect behaviour of the student. Owing to its prototypical and experimental character, ELISA was not ready for concrete, large scale experimentation on any pair of the languages...

Báo cáo khoa học: "Learning to Extract Relations from the Web using Minimal Supervision" ppt

... appropriate for trainingdatasets that contain just a few, very large bags. In a multi-instance kernel approach, only bags (and not instances) are considered as training examples,577which means that ... this may be anoverestimate (w may appear in a sentence contain-ing a due to causes other than a) , and also becauseof data sparsity, the quantity τ(w) may sometimesresult in a negative value ... OpenNLP1package. Sentences that do not contain both arguments a 1 and a 2are discarded. For every remaining sentence,we ﬁnd the occurrences of a 1 and a 2that are clos-est to each other, and create...

Báo cáo khoa học: "Learning Phrase-Based Spelling Error Models from Clickthrough Data" pot

... sen-tences, making a grammar-based approach inap-propriate. Most importantly, many queries con-tain search terms, such as proper nouns and names, which are not well established in the language. For ... q=harrypotter+sheme+park&aq=f&oq=&aqi= http://www.google.com/search? hl=en&ei=rnNAS8-oKsWe_AaB2eHlCA& sa=X&oi=spell&resnum=0&ct= result&cd=1&ved=0CA4QBSgA& q=harry+potter+theme+park&spell=1 ... in training data. These estimates are useful for contextual lexical selection with suffi-cient training data, but can be subject to data sparsity issues. An alternate translation probability...

Báo cáo khoa học: "Learning to Tell Tales: A Data-driven Approach to Story Generation" doc

... of background knowledge contain-ing information about the story plot and its characters. This information is detailed and usually hand crafted. In this paper wepropose a data- driven approach ... of learning methods for re-alizing each of these tasks automatically with-out much hand coding. For example, Duboue and McKeown (2002) and Barzilay and Lapata (2005)propose to learn a content ... (K-F-NSAMP), familiarity(FAM), concreteness (CONC), imagery (IMAG),age of acquisition (AOA), and meaningfulness(MEANC and MEANP).Correlation analysis was used to assess the de-gree of linear relationship...

Tài liệu Báo cáo khoa học: "Finding Hedges by Chasing Weasels: Hedge Detection Using Wikipedia Tags and Shallow Linguistic Features" doc

... corpus statistics and syntactic pat-terns. We take Wikipedia as an alreadyannotated corpus using its tagged weaselwords which mark sentences and phrasesas non-factual. We evaluate the quality ... spe-ciﬁc weasel tag, so that Wikipedia can be viewedas a readily annotated corpus. Based on this data, we have built a system to detect sentences thatcontain linguistic hedges. We compare a base-line ... the same for all annotators toenable measuring agreement. The annotators la-beled the data independently and following anno-tation guidelines which were mainly adopted fromthe Wikipedia style...

Báo cáo khoa học: Dissociation/association properties of a dodecameric cyclomaltodextrinase Effects of pH and salt concentration on the oligomeric state pot

... 5¢-AGTACATGTGGGACGTCACCATGGAGTATGTCCC-3¢ (forward) and 5¢-GGGACATACTC CATGGTGACGTCCCACATGTACT-3¢ (reverse);for the H89V mutant, 5¢-TCTGCTGCAGCA GGGTGTTGAGAAGCGCTGGATG-3¢ (forward) and 5¢-CATCCAGCGCTTCTCAACACCCT ... indicatedthat separate dimers could form a dodecamer and thatFig. 1. Apparent molecular mass of CDase I-5 at various pH valuesdetermined by analytical ultracentrifugation analysis.Dynamics ... enzyme.Sedimentation equilibrium and velocity analyticalultracentrifugationSedimentation equilibrium analytical ultracentrifugationwas performed using a Beckman Optima XL -A analyticalultracentrifuge...

Xem thêm