Datasets:

yxchar

/

citation_intent-tlm

like 2

This is because the binary structure has been verified to be very effective for tree-based translation ( Wang et al. , 2007 ; Zhang et al. , 2011a ) .

There is a general consensus among theoretical linguists that the proper representation of verbal argument structure is event structure -- representations grounded in a theory of events that decompose semantic roles in terms of primitive predicates representing concepts such as causality and inchoativity ( Dowty , 1979 ; Jackendoff , 1983 ; Pustejovsky , 1991b ; Rappaport Hovav and Levin , 1998 ) .

measure the standard intrinsic parser metrics unlabeled attachment score ( UAS ) and labeled attachment score ( LAS ) ( Buchholz and Marsi , 2006 ) .

The EDR has close ties to the named entity recognition ( NER ) and coreference resolution tasks , which have been the focus of several recent investigations ( Bikel et al. , 1997 ; Miller et al. , 1998 ; Borthwick , 1999 ; Mikheev et al. , 1999 ; Soon et al. , 2001 ; Ng and Cardie , 2002 ; Florian et al. , 2004 ) , and have been at the center of evaluations such as : MUC-6 , MUC-7 , and the CoNLL '02 and CoNLL '03 shared tasks .

Accordingly , we convert examples such as ( 27 ) into their generalized equivalents , as in ( 28 ) : ( 28 ) <DET> good man : bon homme That is , where Block ( 2000 ) substitutes variables for various words in his templates , we replace certain lexical items with their marker tag .

How it is done is beyond the scope of this paper but is explained in detail in Fink ( 1983 ) .

These tools use a highly optimised GIS implementation and provide sophisticated Gaussian smoothing ( Chen and Rosenfeld , 1999 ) .

Also relevant is work on the general problems of dialog-act tagging ( Stolcke et al. , 2000 ) , citation analysis ( Lehnert et al. , 1990 ) , and computational rhetorical analysis ( Marcu , 2000 ; Teufel and Moens , 2002 ) .

The inference rules that were necessary to convert one list of properties into another do not sit comfortably within the received NLG pipeline model ( e.g. , Reiter and Dale 2000 ) .

Niyogi ( 2001 ) has developed an agenda-driven chart parser for the feature-driven formalism described above ; please refer to his paper for a description of the parsing algorithm .

An example of psycholinguistically oriented research work can be found in Bond and Hayes ( 1983 ) .

In addition to a referring function , noun phrases ( NP ) can also serve communicative goals such as providing new information about the referent and expressing the speaker 's emotional attitude towards the referent ( Appelt , 1985 ; Jordan , 2000 ) .

Some well-known approaches include rule-based models ( Brill and Resnik 1994 ) , backed-off models ( Collins and Brooks 1995 ) , and a maximumentropy model ( Ratnaparkhi 1998 ) .

There is some literature on procedure acquisition such as the LISP synthesis work described in Biermann et al. ( 1984 ) and the PROLOG synthesis method of Shapiro ( 1982 ) .

Against the background of a growing interest in multilingual NLP , multilingual anaphora / coreference resolution has gained considerable momentum in recent years ( Aone and McKee 1993 ; Azzam , Humphreys , and Gaizauskas 1998 ; Harabagiu and Maiorano 2000 ; Mitkov and Barbu 2000 ; Mitkov 1999 ; Mitkov and Stys 1997 ; Mitkov , Belguith , and Stys 1998 ) .

The list of semantic relations with which we work is based on extensive literature study ( Barker et al. , 1997a ) .

The research described below is taking place in the context of three collaborative projects ( Boguraev , 1987 ; Russell et al. , 1986 ; Phillips and Thompson , 1986 ) to develop a general-purpose , wide coverage morphological and syntactic analyser for English .

In most cases , the accuracy of parsers degrades when run on out-of-domain data ( Gildea , 2001 ; McClosky et al. , 2006 ; Blitzer et al. , 2006 ; Petrov et al. , 2010 ) .

This paper presents experiments with generative content models for analyzing the discourse structure of medical abstracts , which has been confirmed to follow the four-section pattern discussed above ( Salanger-Meyer , 1990 ) .

Following Lekakos and Giaglis ( 2007 ) , one approach for achieving this objective consists of applying supervised learning , where a winning method is selected for each case in the training set , all the training cases are labeled accordingly , and then the system is trained to predict a winner for unseen cases .

Thus , over the past few years , along with advances in the use of learning and statistical methods for acquisition of full parsers ( Collins , 1997 ; Charniak , 1997a ; Charniak , 1997b ; Ratnaparkhi , 1997 ) , significant progress has been made on the use of statistical learning methods to recognize shallow parsing patterns syntactic phrases or words that participate in a syntactic relationship ( Church , 1988 ; Ramshaw and Marcus , 1995 ; Argamon et al. , 1998 ; Cardie and Pierce , 1998 ; Munoz et al. , 1999 ; Punyakanok and Roth , 2001 ; Buchholz et al. , 1999 ; Tjong Kim Sang and Buchholz , 2000 ) .

Due to their remarkable ability to incorporate context structure information and long distance reordering into the translation process , tree-based translation models have shown promising progress in improving translation quality ( Liu et al. , 2006 , 2009 ; Quirk et al. , 2005 ; Galley et al. , 2004 , 2006 ; Marcu et al. , 2006 ; Shen et al. , 2008 ; Zhang et al. , 2011b ) .

11 Nivre ( 2008 ) reports that non-projective and pseudo-projective algorithms outperform the `` eager '' projective algorithm in MaltParser , but our training data did not contain any non-projective dependencies .

In particular , boosting ( Schapire , 1999 ; Abney et al. , 1999 ) offers the possibility of achieving high accuracy from a collection of classifiers which individually perform quite poorly .

For MT the most commonly used heuristic is called grow diagonal final ( Och and Ney 2003 ) .

ECM-F is an entity-constrained mention Fmeasure ( cfXXX ( Luo et al. , 2004 ) for how ECM-F is computed ) , and ACE-Value is the official ACE evaluation metric .

In order to obtain semantic representations of each word , we apply our previous strategy ( Schone and Jurafsky ( 2000 ) ) .

Finkelstein et al. ( 2002 ) annotated a larger set of word pairs ( 353 ) , too .

A more flexible approach is used by Reiter and Sripada ( 2002 ) , where users can specify boundary values for attributes like rainfall , specifying , for example , rain counts as moderate above 7 mm/h , as heavy above 20 mm/h , and so on .

Furthermore , a number of performance features , largely based on the PARADISE dialogue evaluation scheme ( Walker et al. , 1997 ) , were automatically logged , derived , or manually annotated .

Shaw and Hatzivassiloglou ( 1999 ) propose to generalize the direct evidence method so that it can apply to unseen pairs of adjectives by computing the transitive closure of the ordering relation .

We evaluated our translations with IBM 's BLEU evaluation metric ( Papineni et al. , 2001 ) , using the same evaluation method and reference retranslations that were used for evaluation at HLT Workshop 2002 at CLSP ( Haji 6 et al. , 2002 ) .

Other attempts to address efficiency include the fast Transformation Based Learning ( TBL ) Toolkit ( Ngai and Florian , 2001 ) which dramatically speeds up training TBL systems , and the translation of TBL rules into finite state machines for very fast tagging ( Roche and Schabes , 1997 ) .

PR is closely related to the work of Mann and McCallum ( 2007 , 2008 ) , who concurrently developed the idea of using penalties based on posterior expectations of features to guide semi-supervised learning .

The best results on most of our data were obtained using Hidden Naive Bayes ( HNB ) ( Zhang et al. , 2005 ) .

de URL : http://www.sfs.nphil.uni-tuebingen.de/sfb / b4home.html 1 This is , for example , the case for all proposals working with verbal lexical entries that raise the arguments of a verbal complement ( Hinrichs and Nakazawa 1989 ) that also use lexical rules such as the Complement Extraction Lexical Rule ( Pollard and Sag 1994 ) or the Complement Cliticization Lexical Rule ( Miller and Sag 1993 ) to operate on those raised elements .

A number of proposals in the 1990s deliberately limited the extent to which they relied on domain and/or linguistic knowledge and reported promising results in knowledge-poor operational environments ( Dagan and Itai 1990 , 1991 ; Lappin and Leass 1994 ; Nasukawa 1994 ; Kennedy and Boguraev 1996 ; Williams , Harvey , and Preston 1996 ; Baldwin 1997 ; Mitkov 1996 , 1998b ) .

ASARES has been previously applied to the acquisition of word pairs sharing semantic relations defined in the Generative Lexicon framework ( Pustejovsky , 1995 ) and called qualia relations ( Bouillon et al. , 2001 ) .

mlSystem ruleFeats + atomFeats We augment mlSystem ruleFeats with more features from our previous work ( Markert et al. , 2012 ; Hou et al. , 2013a ; Hou et al. , 2013b ) on bridging anaphora recognition and antecedent selection .

The description of the EAGLE workbench for linguistic engineering ( Baldwin et al. 1997 ) mentions a case normalization module that uses a heuristic in which a capitalized word in an ambiguous position should be rewritten without capitalization if it is found lower-cased in the same document .

An exception is Grefenstette et al. ( 2004 ) , who experimented with determining the political orientation of websites essentially by classifying the concatenation of all the documents found on that site .

Until now , translation models have been evaluated either subjectively ( e.g. White and O'Connell 1993 ) or using relative metrics , such as perplexity with respect to other models ( Brown et al. 1993b ) .

The three preprocessing steps ( tokenization , POS-tagging , lemmatization ) are performed using TreeTagger ( Schmid , 1995 ) .

Many statistical parsers ( Ratnaparkhi , 1999 ; Collins , 1999 ; Charniak , 2001 ) are based on a history-based probability model ( Black et al. , 1993 ) , where the probability of each decision in a parse is conditioned on the previous decisions in the parse .

In informal experiments described elsewhere ( Melamed 1995 ) , I found that the G2 statistic suggested by Dunning ( 1993 ) slightly outperforms 02 .

For right-branching structures , the leftcorner ancestor is the parent , conditioning on which has been found to be beneficial ( Johnson , 1998 ) , as has conditioning on the left-corner child ( Roark and Johnson , 1999 ) .

This strategy is certainly the right one to start out with , since anaphora is always the more typical direction of reference in English prose ( Halliday and Hasan 1976 , p. 329 ) .

This indicates that parse trees are usually not the optimal choice for training tree-based translation models ( Wang et al. , 2010 ) .

In Charniak ( 1996 ) and Krotov et al. ( 1998 ) , it was observed that treebank grammars ( CFGs extracted from treebanks ) are very large and grow with the size of the treebank .

For example , the forward-backward algorithm ( Baum , 1972 ) trains only Hidden Markov Models , while ( Ristad and Yianilos , 1996 ) trains only stochastic edit distance .

ment ( Sarkar and Wintner , 1999 ; Doran et al. , 2000 ; Makino et al. , 1998 ) .

Manning ( 1993 ) argues that , aside from missing domain-specific complementation trends , dictionaries produced by hand will tend to lag behind real language use because of their static nature .

Finally , the Natural Language Toolkit ( NLTK ) is a package of NLP components implemented in Python ( Loper and Bird , 2002 ) .

Politically-oriented text Sentiment analysis has specifically been proposed as a key enabling technology in eRulemaking , allowing the automatic analysis of the opinions that people submit ( Shulman et al. , 2005 ; Cardie et al. , 2006 ; Kwon et al. , 2006 ) .

Burkett and Klein ( 2008 ) and Burkett et al. ( 2010 ) focused on joint parsing and alignment .

But the general outlines are reasonably clear , and we can adapt some of the UDRS ( Reyle 1995 ) work to our own framework .

KUbler , McDonald , and Nivre ( 2009 ) describe a `` typical '' MaltParser model configuration of attributes and features .13 Starting with it , in a series of initial controlled experiments , we settled on using buf [ 0-1 ] + stk [ 0-1 ] for word-forms , and buf [ 0-3 ] + stk [ 0-2 ] for POS tags .

For this evaluation , we randomly selected 50 abstracts with disorders indexed as the main topic from abstracts retrieved using PubMed on the five clinical questions described in Sneiderman et al. ( 2005 ) .

The need for information systems to support physicians at the point of care has been well studied ( Covell et al. , 1985 ; Gorman et al. , 1994 ; Ely et al. , 2005 ) .

This approach is taken in computational syntactic grammars ( e.g. Jensen 1986 ) ; the number of unlikely parses is severely reduced whenever possible , but no attempt is made to define only the so-called grammatical strings of a language .

Previous work on Chinese SRL mainly focused on how to transplant the machine learning methods which has been successful with English , such as Sun and Jurafsky ( 2004 ) , Xue and Palmer ( 2005 ) and Xue ( 2008 ) .

For example , the interaction of lexical rules is explored at run-time , even though the possible interaction can be determined at compile-time given the information available in the lexical rules and the base lexical entries .2 Based on the research results reported in Meurers and Minnen ( 1995 , 1996 ) , we propose a new computational treatment of lexical rules that overcomes these shortcomings and results in a more efficient processing of lexical rules as used in HPSG .

`` Coherence , '' as outlined above , can be understood as a declarative ( or static ) version of marker passing ( Hirst 1987 ; Charniak 1983 ) , with one difference : the activation spreads to theories that share a predicate , not through the IS-A hierarchy , and is limited to elementary facts about predicates appearing in the text .

On small data sets all of the Bayesian estimators strongly outperform EM ( and , to a lesser extent , VB ) with respect to all of our evaluation measures , confirming the results reported in Goldwater and Griffiths ( 2007 ) .

In addition , the advantages of using linguistically annotated data over raw data are well documented ( Mair , 2005 ; Granger and Rayson , 1998 ) .

For the A * algorithm ( Hart et al. 1968 ) as applied to speech recognition , the actual path score is typically augmented with an estimated score for the unseen portion .

Robust natural language understanding in Atlas-Andes is provided by Rosé 's CARMEL system ( Rosé 2000 ) ; it uses the spelling correction algorithm devised by Elmi and Evens ( 1998 ) .

The framework represents a generalization of several predecessor NLG systems based on Meaning-Text Theory : FoG ( Kittredge and Polguere , 1991 ) , LFS ( Iordanskaja et al. , 1992 ) , and JOYCE ( Rambow and Korelsky , 1992 ) .

Nevertheless , Juola ( 1998 , page 23 ) observes that `` a slightly more general mapping , where two adjacent terminal symbols can be merged into a single lexical item ( for example , a word and its case-marking ) , can capture this sort of result quite handily . ''

Other similar approaches include those of Cicekli and G ¨ uvenir ( 1996 ) , McTait and Trujillo ( 1999 ) , Carl ( 1999 ) , and Brown ( 2000 ) , inter alia .

( Watanabe et al. , 2007 ; Chiang et al. , 2008 ; Hopkins and May , 2011 ) proposed other optimization objectives by introducing a margin-based and ranking-based indirect loss functions .

Burkett and Klein ( 2012 ) utilized a transformation-based method to learn a sequence of monolingual tree transformations for translation .

Baseline language model : For P0 we used a trigram with modified kneser-ney smoothing [ Chen and Goodman , 1998 ] , which is still considered one of the best smoothing methods for n-gram language models .

Therefore , we preprocess Ontonote-5 .0 to derive mention heads using Collins head rules ( Collins , 1999 ) with gold constituency parsing information and gold named entity information .

Our most accurate single grammar achieves an F score of 91.6 on the WSJ test set , rivaling discriminative reranking approaches ( Charniak and Johnson , 2005 ) and products of latent variable grammars ( Petrov , 2010 ) , despite being a single generative PCFG .

We have presented an ensemble approach to word sense disambiguation ( Pedersen , 2000 ) where multiple Naive Bayesian classifiers , each based on co -- occurrence features from varying sized windows of context , is shown to perform well on the widely studied nouns interest and line .

Over the last decade there has been a lot of interest in developing tutorial dialogue systems that understand student explanations ( Jordan et al. , 2006 ; Graesser et al. , 1999 ; Aleven et al. , 2001 ; Buckley and Wolska , 2007 ; Nielsen et al. , 2008 ; VanLehn et al. , 2007 ) , because high percentages of selfexplanation and student contentful talk are known to be correlated with better learning in humanhuman tutoring ( Chi et al. , 1994 ; Litman et al. , 2009 ; Purandare and Litman , 2008 ; Steinhauser et al. , 2007 ) .

We shall see this in the next example : two sentences , regarded as a fragment of paragraph , are a variation on a theme by Hobbs ( 1979 ) .

Nevertheless , the full document text is present in most systems , sometimes as the only feature ( Sugiyama and Okumura , 2007 ) and sometimes in combination with others see for instance ( Chen and Martin , 2007 ; Popescu and Magnini , 2007 ) - .

The recognizer for these systems is the SUMMIT system ( Zue et al. 1989 ) , which uses a segmental-based framework and includes an auditory model in the front-end processing .

Some recent GRE algorithms have done away with the separation between content determination and linguistic realization , interleaving the two processes instead ( Stone and Webber 1998 ; Krahmer and Theune 2002 ) .

To sample from our proposal distribution , we use a blocked Gibbs sampler based on the one proposed by Goodman ( 1998 ) and used by Johnson et al. ( 2007 ) that samples entire parse trees .

Semantic construction proceeds from the derived tree ( Gardent and Kallmeyer , 2003 ) rather than -- as is more common in TAG -- from the derivation tree .

It allows the construction of a non-TAL ( Shieber , 1994 ) , ( Harbusch & Poller , 2000 ) .

Following previous work ( e.g. , Soon et al. ( 2001 ) and Ponzetto and Strube ( 2006 ) ) , we generate training instances as follows : a positive instance is created for each anaphoric NP , NPj , and its closest antecedent , NPi ; and a negative instance is created for NPj paired with each of the intervening NPs , NPi +1 , NPi +2 , ... , NPj_1 .

TF-IDF ( term frequency-inverse document frequency ) is one of the widely used feature selection techniques in information retrieval ( Yates and Neto , 1999 ) .

The gap mechanism resembles the Hold register idea of ATNs ( Woods 1970 ) and the treatment of bounded domination metavariables in lexical functional grammars ( LFGs ) ( Bresnan 1982 , p. 235 ff . )

To solve these scaling issues , we implement Online Variational Bayesian Inference ( Hoffman et al. , 2010 ; Hoffman et al. , 2012 ) for our models .