Báo cáo khoa học: "A Machine Learning Approach to Pronoun Resolution in Spoken Dialogue" docx

... A Machine Learning Approach to Pronoun Resolution in Spoken DialogueMichael Strube and Christoph M¨ullerEuropean Media Laboratory GmbHVilla BoschSchloß-Wolfsbrunnenweg ... masculine and feminine pronouns (3mf), thirdperson neuter pronouns (3n), and third person pluralpronouns (3p). Since only 3n-pronouns have non-NP-antecedents, we were mainly interested in im-provements ... and non-NP-antecedents in spoken dialogue in the TRAINS do-main. Byron extends a pronoun resolution al-gorithm (Tetrault, 2001) with semantic ﬁltering,thus enabling it to resolve anaphors with...

Báo cáo khoa học: "A Machine Learning Approach to German Pronoun Resolution" docx

... A Machine Learning Approach to German Pronoun Resolution Beata KouchnirDepartment of Computational LinguisticsTübingen University72074 Tübingen, Germanykouchnir@sfs.uni-tuebingen.deAbstractThis ... presents a novel ensemble learning approach to resolving Germanpronouns. Boosting, the method in question, combines the moderately ac-curate hypotheses of several classifiers to form a highly accurate ... trained on a subset of theWall Street Journal, comprising 21 million tokens.The reported F-measure for pronoun resolution is81.5. However, (Morton, 2000) only attempts to resolve singular pronouns,...

Báo cáo khoa học: "A machine-learning approach to the identification of WH gaps" doc

... paper consists in the de-velopment of a machine- learning approach to theidentification of WH gaps. This approach reducesthe lexical prerequisites for this task, while main-taining a high degree ... A machine- learning approach to the identification of WH gapsDerrick HigginsEducational Testing Servicedchiggin@alumni.uchicago.eduAbstract In this paper, we pursue ... subcategorizationproperties of verbs, identifying gaps could also aid in automatic lexical acquisition techniques. Manyother applications are imaginable as well, usingthe gap location to inform intonation, semantics,collocation...

Báo cáo khoa học: "A machine learning approach to the automatic evaluation of machine translation" potx

... a machine learning approach to evaluating the well-formedness of output of a machine translation system, using classifiers thatlearn to distinguish human referencetranslations from machine ... minus number of left-branchingnodes¾ branching index for NPs only¾ branching weight index: number oftokens covered by right-branchingnodes minus number of tokens coveredby left-branching ... A machine learning approach to the automatic evaluation of machine translationSimon Corston-Oliver, Michael Gamon and Chris BrockettMicrosoft...

Báo cáo khoa học: "A Machine Learning Approach to Extract Temporal Information from Texts in Swedish and Generate Animated 3D Scenes" docx

... with alarger training set. It would result in a better over-all performance. Switching from decision trees to other training methods such as Support Vector Ma-chines or using semantically ... for instance to mark polarity by linking anot to its main verb. We identify these links simul-taneously with the event detection. We augmentedthe phrase-structure rules to handle subordinationcases ... small do-main ontology and inserts them into the template.We use the event relations resulting from temporalinformation extraction module to order them. Forall pairs of events in the template,...

Báo cáo khoa học: "A Supervised Learning Approach to Automatic Synonym Identiﬁcation based on Distributional Features" ppt

... similarity set-ting, in contrast, the contributions of context typesare uniformly fixed. In order to elucidate what ishappening in this situation, the investigations on ma-chine learning settings, as ... approach is enhancedwith the supervised machine learning. In this paper, we propose a novel approach to automatic synonym identification based on super-vised learning technique. Firstly, we re-formalizesynonym ... divided into five par-titions to conduct five-fold cross validation, of whichfour partitions were used for learning and the one fortesting. SVMlightwas adopted for machine learn-ing, and...

Báo cáo khoa học: "A Meta Learning Approach to Grammatical Error Correction" doc

... training and test data. We hold out 10% of the data for the test. We applied 20% under-sampling to the training instances that do not have any errors to alleviate data imbalance in the training ... structural learning. We adopted the feature set investigated in De Felice (2008) for article error correction. We use the Stanford coreNLP toolkit1 (Toutanova and Manning, 2000; Klein and Manning, ... structural learning with the common structure obtained from the news corpus. We trained three classifiers for comparison; (1) the classifier (INTEG) trained with the integrated training set...

Tài liệu Báo cáo khoa học: "A Fully Bayesian Approach to Unsupervised Part-of-Speech Tagging∗" docx

... 2005), up to 14 percentage points better than MLE. Weﬁnd improvements both when training fromdata alone, and using a tagging dictionary.1 IntroductionUnsupervised learning of linguistic structure ... dic-tionary information, and Section 6 concludes.2 A Bayesian HMM2.1 Motivation In model-based approaches to unsupervised lan-guage learning, the problem is formulated in termsof identifying latent ... data is obtainedby integrating over all possible values of θ:P (t|w) =P (t|w, θ)P (θ|w)dθ. (3)This distribution can be used in various ways, in- cluding choosing the MAP assignment to the latentvariables,...

Báo cáo khoa học: "A Two-step Approach to Sentence Compression of Spoken Utterances" pdf

... the result, suggestingthe effectiveness of the new introduced features.Regarding the two training settings in reranking,we ﬁnd that there is no gain from reranking whenusing only one best ... andLapata, 2008) investigates sentence compression in broadcast news using an integer linear programming approach. There is only a few existing work in spon-taneous speech domains. (Liu and Liu, ... we167introduced are:• the distance to the next same word and the nextsame POS tag.• a binary feature to indicate if there is a ﬁlledpause or incomplete word in the following 4-word window....

Báo cáo khoa học: "A RULE-BASED APPROACH TO EVALUATING IMPORTANCE IN DESCRIPTIVE TEXTS" pptx

... 1983, and 1984) in which the process of representing the meaning of a natural language text is split into three main tasks, namely: sentence understanding, structure capturing, and i mportance ... values (integers) according to their relative importance in the text. Once the HPN representation of a text has been produced, it is easy to prune the less relevant parts in order to obtain the ... importance values to the different parts of a text and can resolve or explain conflicting evaluations seems more appropriate. Such an approach allows taking into account in a flexible and...

Xem thêm