Papers reviewed and determined not to be word norm studies. Use the flag icon to report errors or suggest re-inclusion.
18265 papers
We propose the use of Lexicalized Tree Adjoining Grammar (LTAG) as a source of features that are useful for reranking the output of a statistical parser. In this paper, we extend the notion of a tree kernel over arbitrary sub-trees of the parse to the derivation trees and derived trees provided by the LTAG formalism, and in addition, we extend the original definition of the tree kernel, making it more lexicalized and more compact. We use LTAG based features for the parse reranking task and obtain labeled recall and precision of 89.7%/90.0% on WSJ section 23 of Penn Treebank for sentences of length ≤ 100 words. Our results show that the use of LTAG based tree kernel gives rise to a 17% relative difference in f-score improvement over the use of a linear kernel without LTAG based features.
Intensified aquaculture has strong impact on fish health by stress and infectious diseases and has stimulated the interest in the orchestration of cytokines and growth factors, particularly their influence by environmental factors, however, only scarce data are available on the GH/IGF-system, central physiological system for development and tissue shaping. Most recently, the capability of the host to cope with tissue damage has been postulated as critical for survival. Thus, the present study assessed the combined impacts of estrogens and bacterial infection on the insulin-like growth factors (IGF) and tumor-necrosis factor (TNF)-α. Juvenile rainbow trout were exposed to 2 different concentrations of 17β-estradiol (E2) and infected with Yersinia ruckeri. Gene expressions of IGF-I, IGF-II and TNF-α were measured in liver, head kidney and spleen and all 4 estrogen receptors (ERα1, ERα2, ERβ1 and ERβ2) known in rainbow trout were measured in liver. After 5 weeks of E2 treatment, hepatic up-regulation of ERα1 and ERα2, but down-regulation of ERß1 and ERß2 were observed in those groups receiving E2-enriched food. In liver, the results further indicate a suppressive effect of Yersinia-infection regardless of E2-treatment on day 3, but not of E2-treatment on IGF-I whilst TNF-α gene expression was not influenced by Yersinia-infection but was reduced after 5 weeks of E2-treatment. In spleen, the results show a stimulatory effect of Yersinia-infection, but not of E2-treatment on both, IGF-I and TNF-α gene expressions. In head kidney, E2 strongly suppressed both, IGF-I and TNF-α. To summarise, the treatment effects were tissue- and treatment-specific and point to a relevant role of IGF-I in infection.
The question of how treebank annotation schemes should be related to linguistic theories has been debated as long as treebanks have existed. Historically speaking, it is probably true to say that there has been a development from mostly theoryneutral annotation schemes to more theoretically oriented frameworks or even annotation
The paper presents a maximum entropy Chinese character-based parser trained on the Chinese Treebank ("CTB" henceforth). Word-based parse trees in CTB are first converted into character-based trees, where word-level part-of-speech (POS) tags become constituent labels and character-level tags are derived from word-level POS tags. A maximum entropy parser is then trained on the character-based corpus. The parser does word-segmentation, POS-tagging and parsing in a unified framework. An average label F-measure 81.4% and word-segmentation F-measure 96.0% are achieved by the parser. Our results show that word-level POS tags can improve significantly word-segmentation, but higher-level syntactic strutures are of little use to word segmentation in the maximum entropy parser. A word-dictionary helps to improve both word-segmentation and parsing accuracy.
Abstract. The problem of Prepositional Phrase (PP) attachment disambiguation consists in determining if a PP is part of a noun phrase, as in He sees the room with books, or an argument of a verb, as in He fills the room with books. Volk has proposed two variants of a method that queries an Internet search engine to find the most probable attachment variant. In this paper we apply the latest variant of Volk’s method to Spanish with several differences that allow us to attain a better performance close to that of statistical methods using treebanks. 1
this paper I will discuss a framework for semantics which allows us to record truth-conditional and compositional analyses as dependency-style corpus annotations in a direct and fine-grained fashion. This method eliminates the need for a semantic representation formalism by decomposing semantic information into simple statements about (word or morpheme) tokens. A collection of such data would form a new kind of linguistic treebank. The main purpose of this article is to show that the present approach makes it possible to combine formal semantics and corpus-oriented study of language use in new and interesting ways. The methodology of this framework, which I call Token Dependency Semantics (TDS, Dahllf [4]), is in several respects different from the common one(s) in traditional formal semantics. TDS nevertheless delivers a fairly conventional (but ontologically restrained) analysis of truth-conditional meaning
This paper describes log-linear parsing models for Combinatory Categorial Grammar (CCG). Log-linear models can easily encode the long-range dependencies inherent in coordination and extraction phenomena, which CCG was designed to handle. Log-linear models have previously been applied to statistical parsing, under the assumption that all possible parses for a sentence can be enumerated. Enumerating all parses is infeasible for large grammars; however, dynamic programming over a packed chart can be used to efficiently estimate the model parameters. We describe a parellelised implementation which runs on a Beowulf cluster and allows the complete WSJ Penn Treebank to be used for estimation.
This paper reports on experiments in classifying the semantic role annotations assigned to prepositional phrases in both the Penn Treebank and FrameNet. In both cases, experiments are done to see how the prepositions can be classified given the dataset's role inventory, using standard word-sense disambiguation features. In addition to using traditional word collocations, the experiments incorporate class-based collocations in the form of WordNet hypernyms. For Treebank, the word collocations achieve slightly better performance: 78.5% versus 77.4% when separate classifiers are used per preposition. When using a single classifier for all of the prepositions together, the combined approach yields a significant gain at 85.8% accuracy versus 81.3% for word-only collocations. For FrameNet, the combined use of both collocation types achieves better performance for the individual classifiers: 70.3% versus 68.5%. However, classification using a single classifier is not effective due to confusion among the fine-grained roles.
This paper reports on the use of two distinct evaluation metrics for assessing a stochastic parsing model consisting of a broad-coverage Lexical-Functional Grammar (LFG), an efficient constraint-based parser and a stochastic disambiguation model. The first evaluation metric measures matches of predicate-argument relations in LFG f-structures (henceforth the LFG annotation scheme) to a gold standard of manually annotated f-structures for a subset of the UPenn Wall Street Journal treebank. The other metric maps predicate-argument relations in LFG f-structures to dependency relations (henceforth DR annotations) as proposed by Carroll et al. (Carroll et al., 1999). For evaluation, these relations are matched against Carroll et al.'s gold standard which was manually annnotated on a subset of the Brown corpus. The parser plus stochastic disambiguator gives an F-measure of 79% (LFG) or 73% (DR) on the WSJ test set. This shows that the two evaluation schemes are similar in spirit, although accuracy is impaired systematically by mapping one annotation scheme to the other. A systematic loss of accuracy is incurred also by corpus variation: Training the stochastic disambiguation model on WSJ data and testing on Carroll et al.'s Brown corpus data yields an F-score of 74% (DR) for dependency-relation match. A variant of this measure comparable to the measure reported by Carroll et al. yields an F-measure of 76%. We examine divergences between annotation schemes aiming at a future improvement of methods for assessing parser quality.
▪ Abstract Using Australian languages as examples, cultural selection is shown to shape linguistic structure through invisible hand processes that pattern the unintended outcomes (structures in the system of shared linguistic norms) of intentional actions (particular utterances by individual agents). Examples of the emergence of culturally patterned structure through use are drawn from various levels: the semantics of the lexicon, grammaticalized kin-related categories, and culture-specific organizations of sociolinguistic diversity, such as moiety lects, “mother-in-law” registers, and triangular kin terms. These phenomena result from a complex of diachronic processes that adapt linguistic structures to culture-specific concepts and practices, such as ritualization and phonetic reduction of frequently used sequences, the input of shared cultural knowledge into pragmatic interpretation, semanticization of originally context-dependent inferences, and the input of linguistic ideologies into the systematization of lectal variants. Some of these processes, such as the emergence of subsection terminology and moiety lects, operate over speech communities that transcend any single language and can only be explained if the relevant processes take the multilingual speech community as their domain of operation. Taken together, the cases considered here provide strong evidence against nativist assumptions that see linguistic structures simply as instantiations of biologically given “mentalese” concepts already present in the mind of every child and give evidence in favor of a view that sees individual language structures as also conditioned by historical processes, of which functional adaptation of various kinds is most important. They also illustrate how, in the domain of language, stable socially shared structures can emerge from the summed effects of many communicative micro-events by individual agents.
We present a neural network method for inducing representations of parse histories and using these history representations to estimate the probabilities needed by a statistical left-corner parser. The resulting statistical parser achieves performance (89.1% F-measure) on the Penn Treebank which is only 0.6% below the best current parser for this task, despite using a smaller vocabulary size and less prior linguistic knowledge. Crucial to this success is the use of structurally determined soft biases in inducing the representation of the parse history, and no use of hard independence assumptions.
Perceptions of social closeness and familiarity were assessed among 44 monozygotic (MZA) and 33 dizygotic (DZA) reunited twin pairs, and several individual twins and triplets. Significantly greater MZA than DZA closeness and familiarity were found. Closeness and familiarity ratings for co-twins exceeded those for nonbiological siblings with whom twins were raised. Correlations between perceptions of physical resemblance and social closeness and familiarity were positive and statistically significant. However, most correlations between social relatedness and contact time were non-significant. Associations between social relatedness and similarities in selected behavioral traits were also examined. The findings support various theoretical perspectives anticipating greater affiliation among close relatives than distant relatives.
A semantic database has been extended with visual information to enable video annotation. This paper describes a lexical database, WordNet. We show its limitations with respect to describing visual characteristics, and describe an extension to WordNet that contains specific visual information. Having such a semantic database makes video annotation possible for broadcast news: a domain that can cover any topic and involve a wide variety of events, objects and scenes. Combining basic visual analysis techniques and a semantic database containing visual descriptions avoids the problem developing large numbers of specific object and event detectors. Such a semantic database can be of great value for the analysis of multi-modal information. As far as we know, such a database has not been developed before.
INTRODUCTION: Although evidence suggests that interpersonal psychotherapy may be an efficacious treatment for eating disorders, there is surprisingly little systematic knowledge about the interpersonal world of these patients. METHOD: SASB self-image ratings were used to explore interpersonal profiles in a large heterogeneous sample of eating disorders (N = 830), matched normal controls (N = 105) and a small group of controls with subclinical depression (N = 26). RESULTS: Eating disorder patients clearly presented with significantly more negative interpersonal profiles compared to controls. Within the eating disorder group, anorexics were characterized by high self-control, self-blame and self-attack. Patients with binge eating disorder expressed the least negative self-image, and were significantly more self-affirming than bulimics and less self-controlling than patients with atypical eating disorders. CONCLUSIONS: Eating disorder patients may have distinct interpersonal profiles that increase the risk of negative therapeutic reaction. Better knowledge of interpersonal processes in eating disorders may help to improve both diagnostic assessment and treatment.
XML and related W3C standards (XSLT, XML Schema, XPath, DOM, etc.) often take part in the linguistic data representation and interchange today but not as a direct way to implement efficient data manipulation software tools. The aim of the paper is to show that incorporation of XML and the standards that surround it can bring general applicability of the implemented system for various kinds of linguistic data and also easy extensibility of such systems. As a case study we present a designed and implemented system called DEB (Dictionary Editor and Browser) that is able to manage lexical data from dictionaries to lexical databases, semantic networks, and complex ontologies. The smart design also facilitates the connection to other linguistic tools such as corpus managers or morphological analysers.
Past research has found that individual differences in both attitudinal and situational variables may be associated with males’ likelihood of acquaintance rape (LAR). The present research was conducted to examine the predictive value of both attitudinal and situational factors on males’ likelihood of forcing a female acquaintance to have non-consensual sexual intercourse. In Study 1, male and female respondents (Rs) were presented with a scenario depicting a hypothetical sexual interaction between the respondent and a newly acquainted member of the opposite sex. As the encounter progressed from one sexual activity to the next, Rs made three ratings regarding their own and partner’s intent to engage in each successive activity. The scenario ended with the female refusing further activity and males’ affect ratings, adherence to attitudes conducive of rape, and LAR were measured. Males’ initial perceptions of female sexual intent (to later engage in sexual intercourse) best predicted LAR. Study 2 was conducted to examine the role of female sexual communication on perceptions of consent to sexual intercourse. Rs were presented with a scenario similar to that in Study 1, but at each stage were requested to rate the extent to which the female had consented to engage in each sexual activity. Males completed the same affect and attitudinal measures. The results of Study 2 again suggested that males’ initial perception of female consent to (later) engage in sexual intercourse best predicted LAR. The present research suggests that further investigation into the role of situational factors and males’ initial perception of sexual intent and consent in the aetiology of acquaintance rape is required.
This paper presents a Java-based hyperbolic-style browser designed to render RDF files as structured ontological maps. The program was motivated by the need to browse the content of a web-accessible ontology server: WEB KB-2. The ontology server contains descriptions of over 74,500 object types derived from the WordNet 1.7 lexical database and can be accessed using RDF syntax. Such a structure creates complications for hyperbolic-style displays. In WEB KB-2 there are 140 stable ontology link types and a hyperbolic display needs to filter and iconify the view so different link relations can be distinguished in multi-link views. Our browsing tool, OntoRama, is therefore motivated by two possibly interfering aims: the first to display up to 10 times the number of nodes in a hyperbolic-style view than using a conventional graphics display; secondly, to render the ontology with multiple links comprehensible in that view.
This paper investigates adapting a lexicalized probabilistic context-free grammar (PCFG) to a novel domain, using maximum a posteriori (MAP) estimation. The MAP framework is general enough to include some previous model adaptation approaches, such as corpus mixing in Gildea ( Other approaches falling within this framework are more effective. In contrast to the results in Gildea ( MAP adaptation can also be based on either supervised or unsupervised adaptation data. Even when no in-domain treebank is available, unsupervised techniques provide a substantial accuracy gain over unadapted grammars, as much as nearly 5% F-measure improvement.
We use the grammatical relations (GRs) described in Carroll et al. (1998) to compare a number of parsing algorithms. A first ranking of the parsers is provided by comparing the extracted GRs to a gold standard GR annotation of 500 Susanne sentences: this required an implementation of GR extraction software for Penn Treebank style parsers. In addition, we perform an experiment using the extracted GRs as input to the Lappin and Leass (1994) anaphora resolution algorithm. This produces a second ranking of the parsers, and we investigate the number of errors that are caused by the incorrect 'GRs.
Digitizing and annotating texts and field recordings Given that several initiatives worldwide currently explore the new field of documentation of endangered languages, the E-MELD project proposes to survey and unite procedures, techniques and results in order to achieve its main goal, ''the formulation and promulgation of best practice in linguistic markup of texts and lexicons''. In this context, this year's workshop deals with the processing of recorded texts. I assume the most valuable contribution I could make to the workshop is to show the procedures and methods used in the Awetí Language Documentation Project. The procedures applied in the Awetí Project are not necessarily representative of all the projects in the DOBES program, and they may very well fall short in several respects of being best practice, but I hope they might provide a good and concrete starting point for comparison, criticism and further discussion. The procedures to be exposed include: * taping with digital devices, * digitizing (preliminarily in the field, later definitely by the TIDEL-team at the Max Planck Institute in Nijmegen), * segmenting and transcribing, using the transcriber computer program, * translating (on paper, or while transcribing), * adding more specific annotation, using the Shoebox program, * converting the annotation to the ELAN-format developed by the TIDEL-team, and doing annotation with ELAN. Focus will be on the different types of annotation. Especially, I will present, justify and discuss Advanced Glossing, a text annotation format developed by H.-H. Lieb and myself designed for language documentation. It will be shown how Advanced Glossing can be applied using the Shoebox program. The Shoebox setup used in the Awetí Project will be shown in greater detail, including lexical databases and semi-automatic interaction between different database types (jumping, interlinearization). ( Freie Universität Berlin and Museu Paraense Emílio Goeldi, with funding from the Volkswagen Foundation.)
We describe an algorithm for recovering non-local dependencies in syntactic dependency structures. The pattern-matching approach proposed by Johnson (2002) for a similar task for phrase structure trees is extended with machine learning techniques. The algorithm is essentially a classifier that predicts a non-local dependency given a connected fragment of a dependency structure and a set of structural features for this fragment. Evaluating the algorithm on the Penn Treebank shows an improvement of both precision and recall, compared to the results presented in (Johnson, 2002).
We investigate the performance of the Structured Language Model when one of its components is modeled by a connectionist model. Using a connectionist model and a distributed representation of the items in the history makes the component able to use much longer contexts than possible with currently used interpolated or backoff models, both because of the inherent capability of the connectionist model to fight the data sparseness problem, and because of the only sub-linear growth in the model size when increasing the context length. Experiments show significant improvement in perplexity and moderate reduction in word error rate over the baseline SLM results on the UPENN treebank and Wall Street Journal (WSJ) corpora respectively. The results also show that the probability distribution obtained by our model is much less correlated to regular N-grams than the baseline SLM model.
El estudio experimental de la emoción requiere de estímulos que evoquen en una forma confiable reacciones psicológicas y fisiológicas que varien sistemáticamente sobre el rango de emociones de acuerdo a las dimensiones de valencia (agradable o desagradable), activación (excitado o calmado) y dominancia (alta y baja) (Lang, Bradley, Cuthbert, 1999). A pesar de que los correlatos neurales de las emociones básicas han sido investigados, la organización neural de las "emociones morales" en el cerebro humano no se conocen bien. El objetivo de la presente investigación fue obtener un grupo de estimulos diferenciados (fotografías) y caracterizarlos en términos de su valencia afectiva, activación, dominancia, y contenido moral, en una población mexicana. Se seleccionaron fotografías que representan escenas con una carga emocional amplia como violaciones morales (escenas de guerra, asaltos físicos, etc), escenas aversivas sin connotación moral (tumores, cuerpos mutilados) y escenas naturales (toallas, mesas, puertas, etc. ). Los sujetos evaluaron cada fotografía de acuerdo a su valencia, activación, dominancia y contenido moral (ausente o extremo). Para la evaluación, se utilizó la Escala Internacional Self-Assessment Maniki Affective Rating System desarrollada por Lang (1980). Se discute las implicaciones de los datos, para el estudio de las emociones y del juicio moral.
Dealing with convergence in German speech islands in Russia, Brazil and the United states the article discusses the linguistic phenomena related to the notion of convergence from different vantage points including intralinguistic convergence (due to dialect-dialect contact), interlinguistic convergence (due to language-language contact), typological "convergence" (or intralinguistic change), pidginization, and cognitive processes of simplification. Most of the German speech islands are considered to be contracting - if not dying - varieties with respect to the reduction of their grammatical systems. Evidently, for a long time language contact (and sometimes variety contact) have severely increased. Linguistic norms have been weakened in terms of both norm certainty and norm loyalty thus giving way to processes similar to those common to pidgin languages. External induced changes are highly remarkable in all German speech islands. But the susceptibility for change and the ways of change are structured by systematical and typological constraints which probably turn out to be cognitive processes underlying quite "normal" linguistic change. This change is discussed as a subsequent process of "regularization" (of irregular forms), simplification (of rules) and loss of grammatical distinctions (and their compensation). The linguistic description of these interrelated processes is based on an integrated approach providing methodology from sociolinguistics, dialectology and research on language change, including the attempt to highlight the cognitive structures which furrow the line for internal simplifications under external pressure. Comparative speech island research seems to be a promising field of application for the description of the intermesh of these processes.
This article describes three statistical models for natural language parsing. The models extend methods from probabilistic context-free grammars to lexicalized grammars, leading to approaches in which a parse tree is represented as the sequence of decisions corresponding to a head-centered, top-down derivation of the tree. Independence assumptions then lead to parameters that encode the X-bar schema, subcategorization, ordering of complements, placement of adjuncts, bigram lexical dependencies, wh-movement, and preferences for close attachment. All of these preferences are expressed by probabilities conditioned on lexical heads. The models are evaluated on the Penn Wall Street Journal Treebank, showing that their accuracy is competitive with other models in the literature. To gain a better understanding of the models, we also give results on different constituent types, as well as a breakdown of precision/recall results in recovering various types of dependencies. We analyze various characteristics of the models through experiments on parsing accuracy, by collecting frequencies of various structures in the treebank, and through linguistically motivated examples. Finally, we compare the models to others that have been applied to parsing the treebank, aiming to give some explanation of the difference in performance of the various models.
Acronyms are a very dynamic area of the lexicon of many languages. A hybrid, modular methodology for the acquisition of acronyms is presented, which uses an existing acronym-expansion matching component, and machine learning in two separate phases for the identification of long-distance acronym definition patterns.The resulting system, using Support Vector Machines (SVM) is trained on 600 news stories from the Wall Street Journal component of the Penn Treebank corpus using a number of lexical, syntactic, and acronym-expansion matching features. Statistical cooccurrence information for acronym-expansion pairs is extracted from search engine hit counts.The system achieves Fβ=1=92.38% on 400 news stories from the same source and has good asymptotic efficiency, making it adequate for the automatic extraction of acronyms even from noisy sources, such as newspaper text.
We present a probabilistic parsing model for German trained on the Negra treebank. We observe that existing lexicalized parsing models using head-head dependencies, while successful for English, fail to outperform an unlexicalized baseline model for German. Learning curves show that this effect is not due to lack of training data. We propose an alternative model that uses sister-head dependencies instead of head-head dependencies. This model outperforms the baseline, achieving a labeled precision and recall of up to 74%. This indicates that sister-head dependencies are more appropriate for treebanks with very flat structures such as Negra. 1
The primary purpose of this study was to assess the cross-cultural invariance of job performance ratings. A secondary purpose was to examine potential cross-cultural differences in correlates of performance ratings (i.e., ratee sex, age, tenure; supervisor's opportunity to observe ratee). Fast-food supervisors from Canada, South Korea, and Spain rated employees on their technical proficiency, customer service, and teamwork. Results show that these ratings demonstrate a basic level of measurement invariance, although the error variances of the ratings and pattern of construct variances and covariances were largely culture-specific. This suggests that supervisors across cultures may use and interpret the ratings similarly, but perceive differences in performance. Furthermore, age, tenure, and the supervisor's opportunity to observe the ratee were found to affect ratings differently across cultures. Overall, this study suggests that although job performance ratings are at least partially invariant across cultures, latent performance may not be, and we present some preliminary data as to why latent invariance may not exist.
In this paper we will present work carried out lately on the 50,000 words Italian Spontaneous Speech Corpus called AVIP, under national project API, made available for free download from the website of the coordinator, the University of Naples. We will concentrate on the tuning of the parser for Italian which had been previously used to parse 100,000 words corpus of written Italian within the National Treebank initiative coordinated by ILC in Pisa. The parser receives as input the adequately transformed orthographic transcription of the dialogues making up the corpus, in which pauses, hesitations and other disfluencies have been turned into most likely corresponding punctiation marks, interjections or truncation of the word underlying the uttered segment.\nThe most interesting phenomenon we will discuss is without any doubts "overlapping", i.e. a speech event in which two people speak at the same time by uttering actual words or in some cases nonwords, when one of the speakers, usually the one which is not the current turntaker, interrupts the current speaker.\nThis phenomenon takes place at a certain point in time where it has to be anchored to the speech signal but in order to be fully parsed and subsequently semantically interpreted, it needs to be referred semantically to a following turn.
We have developed an example-based machine translation (EBMT) system that uses the World Wide Web for two different purposes: First, we populate the system's memory with translations gathered from rule-based MT systems located on the Web. The source strings input to these systems were extracted automatically from an extremely small subset of the rule types in the Penn-II Treebank. In subsequent stages, the source, target translation pairs obtained are automatically transformed into a series of resources that render the translation process more successful. Despite the fact that the output from on-line MT systems is often faulty, we demonstrate in a number of experiments that when used to seed the memories of an EBMT system, they can in fact prove useful in generating translations of high quality in a robust fashion. In addition, we demonstrate the relative gain of EBMT in comparison to on-line systems. Second, despite the perception that the documents available on the Web are of questionable quality, we demonstrate in contrast that such resources are extremely useful in automatically postediting translation candidates proposed by our system.
Abstract An elevated disgust sensitivity (DS) is considered to be a vulnerability factor for the development of a blood-injection-injury (BII) phobia. Within the present functional Magnetic Resonance Imaging (fMRI) study, 12 female BII phobics were scanned while viewing alternating blocks of 40 disgust-inducing, 40 fear-inducing, and 40 affectively neutral pictures. Each block lasted 60s and was repeated six times during the experiment. All scenes were phobia-irrelevant. Afterwards, the subjects gave affective ratings for the pictures and described their DS on a self-report measure for different areas (e.g., poor hygiene, unusual food, death/deformation). The responses were compared with those of 12 nonphobic females. The BII phobics showed a stronger occipital activation within the right cuneus and lingual gyrus during the first viewing of the disgusting pictures. Aside from this finding, which could be interpreted as reflecting increased attention, there was little evidence for a generally elevated DS in BII phobia. On the DS questionnaire, the patients had indicated a greater reactivity only for disorder-relevant contents (death/deformation). Further, both groups gave similar disgust ratings for the pictures and showed comparable brain-dynamic responses over all blocks of the disgust condition, which included the activation of both amygdalae and the left inferior frontal gyrus.
Natural language generation (NLG) is the task of formulating a fluent sequence of words in natural language to communicate information or ideas in applications like machine translation, human-computer dialogue, automatic summarization, and question-answering. Realization, a fundamental subtask of NLG, produces an individual sentence from a sentence plan specified in terms of linguistic relations between words and/or concepts. It involves determining the order of words, inserting function words like determiners and prepositions, performing morphological inflections, and ensuring grammaticality and agreement. An ultimate goal for natural language generation is to develop a large-scale, robust, general-purpose system. Two primary challenges are scaling up to broad coverage of syntax and producing high quality output. The irregularity of natural language makes it difficult to know how to combine linguistic primitives into fluent sentences. Also, the knowledge resources for making such a determination are time-consuming and labor-intensive to assemble, leading to a knowledge acquisition bottleneck. Evaluating whether a realizer performed appropriately is an additional challenge. There can often be more than one acceptable output, and no tools exist that can automatically assess grammaticality or fluency. This thesis takes an approach of using probabilistic models learned from text corpora to rank candidate sentences and output the most likely. It contributes (1) a symbolic mapping rule formalism and ruleset for mapping inputs to candidate outputs that achieves broad coverage through greater regularity, (2) a packed forest representation and efficient ranking algorithm that can manage the combinatorial growth in output candidates, and (3) an empirical evaluation of coverage, correctness, and the ability to handle underspecification. This evaluation is the first large-scale empirical evaluation of coverage and quality ever performed for sentence realization. The empirical evaluation is performed by automatically converting a set of 2400 hand-parsed sentences from the Penn Treebank corpus into system inputs, and then regenerating them using the system. The top-ranked output of the generator is compared to the original sentence. The results show better than 80% coverage of newspaper text and 94% precision (57% are exact matches) for almost fully-specified inputs, and the same coverage with 55% precision for minimally specified inputs.
We present a system for automatically identifying PropBank-style semantic roles based on the output of a statistical parser for Combinatory Categorial Grammar. This system performs at least as well as a system based on a traditional Treebank parser, and outperforms it on core argument roles.
This work deals with models used, or usable in the domain of Automatic Natural Language Processing, when one seeks a syntactic interpretation of a statement. This interpretation can be used as additional information for subsequent treatments, that can aim for instance at producing a semantic representation of the statement. It can also be used as a filter to select utterances belonging to a specific language, among several hypotheses, as done in Automatic Speech Recognition. As the syntactic interpretation of a statement is generally ambiguous with natural languages, the probabilisation of the space of syntactic trees can help in the analysis task: when several analyses are competing, one can then extract the most probable interpretation, or classify interpretations according to their probabilities. We are interested here in the probabilistic versions of Context-Free Grammars (PCFGs) and Substitution Tree Grammar (PTSGs). Syntactic treebanks, which as much as possible account for the language we wish to model, serve as the basis for defining the probabilistic parameters of such grammars. First, we exhibit in this thesis some drawbacks of the usual learning paradigms, due to the use of arbitrary heuristics (STSG DOP model), or to the use of learning criteria that consider these grammars as generative ones (creation of sentences from the grammar) rather than dedicated to analysis (creation of analyses from the sentence). In a second time, we propose new methods for training grammars, based on the traditional Maximum Entropy and Maximum Likelihood criteria. These criteria are instanciated so that they correspond to a syntactic analysis task rather than a language generation task. Specific training algorithms are necessary for their implementation, but traditional algorithms can cope with those models for the task of syntactic analysis. Lastly, we invest the problem of time complexity of syntactic analysis, which is a real issue for the effective use of PTSGs. We describe classes of PTSGs that allow the analysis of a sentence in polynomial complexity. We finally describe a method that enable the extraction of such a PTSG from the set of subtrees of a treebank. The PTSG produced by this method allows us to test our non-generative learning criterium on "realistic" data, and to give a statistical comparison between this criterium and the usual heuristic criterium in term of analysis performance.
This paper describes the use of clustering at three stages within a larger research effort to identify semantic frames used in English automatically. The first of two tasks within this effort has been the identification of sets of semantically related verb senses that invoke a common semantic frame. Within this task, clustering has been used both to build sets of verb senses with the potential of invoking a common semantic frame and then to merge sets with a high degree of overlap. The paper is organized as follows: Section 2 introduces frame semantics. Section 3 outlines the methodology used to identify sets of semantically related verb senses that invoke a common semantic frame, while section 4 presents the specific clustering algorithm used within that process. Section 5 discusses the use of this clustering algorithm for the identification of semantically related verbs in two machine-readable lexical resources: the machine-readable version of the Longman Dictionary of Contemporary English (LDOCE, 1978 edition) and WordNet, an online lexical database (http://www.cogsci.princeton.edu/-wn; version 1.7.1 has been used for the work reported here). Section 6 presents the use of clustering to merge overlapping sets of verb senses formed in previous steps. Section 7 discusses the results of these clusterings, paying particular attention to the effect of LDOCE's restricted defining vocabulary on the clustering process.
Choosing the statistical model is the key problem in statistical parsing. Statistical model lies in the core of NLP parsing. This paper investigates 4 primary statistical parsing models, namely PCFG, history-based model, cascading parsing model and head-driven parsing model, and compares their performances in a 10000 Chinese treebank. The analysis based on the experiment were shown in the paper. The comparative study of these models can be exploited to build the practical and effective Chinese parser.
This paper describes a fast algorithm that selects features for conditional maximum entropy modeling. Berger et al. (1996) presents an incremental feature selection (IFS) algorithm, which computes the approximate gains for all candidate features at each selection stage, and is very time-consuming for any problems with large feature spaces. In this new algorithm, instead, we only compute the approximate gains for the top-ranked features based on the models obtained from previous stages. Experiments on WSJ data in Penn Treebank are conducted to show that the new algorithm greatly speeds up the feature selection process while maintaining the same quality of selected features. One variant of this new algorithm with look-ahead functionality is also tested to further confirm the good quality of the selected features. The new algorithm is easy to implement, and given a feature space of size F, it only uses O(F) more space than the original IFS algorithm.
Machine translation engines draw on various types of databases. This paper is concerned with Arabic as a source or target language, and focuses on lexical databases. The non-concatenative nature of Arabic morphology, the complex structure of Arabic word-forms, and the general use of vowel-free writing present a real challenge to NLP developers. We show here how and why a stem-grounded lexical database, the items of which are associated with grammar-lexis specifications – as opposed to a root-&-pattern database –, is motivated both linguistically and with regards to efficiency, economy and modularity. Arguments in favour of databases relying on stems associated with grammar-lexis specifications (such as DIINAR.1 or the Arabic dB under development at SYSTRAN), rather than on roots and patterns, are the following: (a) The latter include huge numbers of rule-generated word-forms, which do not actually appear in the language. (b) Rule-generated lemmas – as opposed to existing ones – are widely under-specified with regards to grammar-lexis relations. (c) In a Semitic language such as Arabic, the mapping of grammar-lexis specifications that need to be associated with every lexical entry of the database is decisive. (d) These specifications can only be included in a stem-based dB. Points (a) to (d) are crucial and in the context of machine translation involving Arabic.
In den letzten Jahren ist die Zahl der verfgbaren linguistisch annotierten Korpora stndig gewachsen. Zu den bekanntesten gehren das Brown-Korpus, das Susanne-Korpus, die Penn-Treebank, das Negra-Korpus, das Tiger-Korpus und die im Zusam-