Papers reviewed and determined not to be word norm studies. Use the flag icon to report errors or suggest re-inclusion.
18625 papers
Abstract This paper describes a framework for building story traces (compact global views of a narrative) and story projections (selections of key elements of a narrative) and their applications in text understanding and classification. Word and sense properties are extracted from documents using the WordNet lexical database enhanced with Prolog inference rules and a number of lexical transformations. Inference rules are based on navigation in various WordNet relation chains (hypernyms, meronyms, entailment and causality links, etc.) and derived relations expressed as Boolean combinations of node and edge properties used to direct the navigation. The resulting abstract story traces provide a compact view of the underlying narrative's key content elements and a means for automated indexing and classification of text collections. Ontology driven projections act as a kind of “semantic lenses” and provide a means to select a subset of a narrative whose key sense elements are subsumed by a set of concepts, predicates and properties expressing the focus of interest of a user. Finally, we discuss applications of these techniques in text understanding, classification of text collections and answering questions about a text.
One of the major problems in the implementation of natural language processing (NLP) or machine translation (MT) is a complete lexicon: the place where the system's information about words is stored. There are difficulties in deciding what information should be stored in a lexicon and even greater difficulties in acquiring this information in proper form. The OriNet system was designed to incorporate a multiple lexical database and tools under one consistent functional interface in order to facilitate systems requiring syntactic, semantic and lexical information of the Oriya language. We divide the work into two independent tasks. One is to write the source file that contains basic lexical data, the content of these files being the lexical substance of OriNet. The second is to create a set of programs that would accept the source files and process them ultimately to display for the user. This paper describes ongoing work on designing an object oriented model for the OriNet system. It uses object oriented programming, particularly the rich library of classes and programming principles which Java offers. It also provides a convenient tool to conceptualise the process of the OriNet system. This technique also allows flexibility and extensibility of the system with more robustness.
Article Sprachwissen im Konflikt. Sprachliche Zweifelsfälle zwischen Linguistik und Sprachnorm. Arbeitsgruppe auf der Jahrestagung der Deutschen Gesellshaft für Sprachwissenschaft 2003 mit dem Rahmenthema Sprache. Wissen. Sprachwissenschaft, München 26.28. Februar 2003 [Languageknowledge in conflict. Borderline cases between linguistics and linguistic norms. Working group at the annual conference of the German Linguistic Society 2003 with the topic Language. Knowledge. Linguistics.] was published on March 25, 2004 in the journal Zeitschrift für germanistische Linguistik (volume 31, issue 2).
This study examines expressional styles and plans for the education of Korean language in computer chatting rooms, which are relevant to hypertext, as a part of preparing the contents and methods of hypertext expression education\n\n First, because of the 'anonymity' of computer chatting rooms, people can expression their feelings without concealing. On the other hand, anonymity causes flaming and undermines linguistic morality. Conversations in computer chatting rooms occur through direct feedback using 'interaction' Because many meetings and conversations in computer chatting rooms are improvised, they are more impersonal than sincere, and politeness and rules necessary for conversation are often ignored Language in chatting rooms is exchanged by texts, but they contain oral elements, which are expressed through the mouth, as well as non-oral elements That is, the language. is mixed with oral words and textual words.\n\n Telecommunication language is used based on strategies for preserving efficiency, those for preserving expressivity, and psychological factors. Although some view it negatively saying that it destroys linguistic norms, its positive side of creative utilization of texts is not negligible, The reason that such telecommunication language is perceived negatively is its influence on everyday language. Concerning the influence, further research is required.\n\n Research related to the education of expression in computer chatting rooms is mostly focused on telecommunication language. As the present researcher investigated, in addition, contents in Writing and Korean Language Life in the 7th Education Curricula also deal mainly with the destruction of linguistic norms by telecommunication language and problems in linguistic morality.\n\n Thus, the researcher proposes plans for the education of Korean Language related to expressions in computer chatting rooms as follows. ① It is necessary to understand telecommunication language positively, regarding it as a social dialect. ② In a sense, abbreviations, emoticons (emotion icons) and symbolic words are language creation, which are necessary for efficient conversations in communication. ③ Except the examples presented in ②, telecommunication language should not be transferred to everyday language, and for this students must learn grammar intensively and have an ability to distinguish virtual worlds from the real world ④ Students must be given not only theoretic education on the characteristics of oral and textual words rot also one related to conversational expressions in computer chatting rooms in 'Speech' and 'Narration' classes.
The aim of this article is to show that the role of legal terminology in juridical discourse can adequately be determined in a framework of a "juridical textwork" model only. This implies to analyse their role under pragmatic and intertextual aspects. lt will be contended that neither semantics nor the wording of the law can be employed as a starting point for an adequate interpretation of legal language in public discourse. Of crucial importance is rather the question of how the legal text and social reality can combine to constitute the legal norm (which is more than just the legal text). This is illustrated with the example of German court decisions on the issue of sit-ins that were organised by the peace movement in the l970s and l980s to block the access to American army bases. lt will be demonstrated that in the process of putting the coercion law in concrete normative terms through different courts, specific legal terms are semanticly modified and adjusted to specific language use, hence constituting "semantic battles" and linguistic norm conflicts in the juridical discourse.
Recovering the semantics, or the meaning, expressed using natural languages is one of the central goals of the field of natural language processing. The task is challenging due to the large number of ambiguities present in natural languages, such as part-of-speech assignments, structural dependencies, and word sense ambiguity. Reliably resolving these ambiguities would allow computers to gain access to the knowledge represented using natural languages, the format most commonly used in human communications. In this thesis we approach this complex problem by first decomposing it into four smaller problems, or subtasks: part of speech (POS) tagging, word sense disambiguation (WSD), chunking, and parsing. With this decomposition each subtask captures a subset of the ambiguities, thus simplifying the problem and facilitating the application of exact algorithms to maximize accuracy. We then integrate the decisions from the subtasks to form the most plausible interpretation, in contrast to top-down modeling that is prone to error propagation. We first apply machine learning algorithms to automatically train probabilistic models based on annotated training corpus. We select a powerful probability model, maximum entropy, to incorporate diverse contexts systematically. We then capture the dependencies between the words within sentences using Bayesian networks, with which we compute disambiguation decisions using an exact inferencing algorithm. To share information between subtasks, we introduce a new structural representation, called Cores-and-Modifiers, to succinctly describe structural features in improving both POS tagging and WSD. We also identify semantic contexts based on the WordNet lexical database to improve both chunking and parsing accuracy. To form the overall interpretation across these subtasks, we introduce an integrative process, instead of a top-down model. Because the entire model is probabilistic, it enables the systematic re-integration of the diverse decisions from the subtasks based on their probabilities. To further improve this integration, we present the concept that the most uncertain decisions are the most error-prone, and thus their alternatives should also be examined. This process, named Most-probable Hypotheses Evaluation, selectively examines a small set of alternate hypotheses to better determine the most plausible interpretation across subtasks, instead of as disparate decisions. The resulting model, named Integrative, Probabilistic Natural-language Parser and Interpreter (IPNPI), integrates the four subtasks to form the most plausible interpretations. The IPNPI model is evaluated by its accuracy on each of the four subtasks using standardized procedures, and we show that it improves the state-of-the-art accuracy in POS tagging, word sense disambiguation, chunking, and parsing. The synthesis of our separate-but-integrative approach is that the IPNPI model is able to accurately and efficiently resolve natural language ambiguities, by producing the most plausible interpretations across part of speech assignment, word sense distinction, phrasal identification, and structural dependencies.
We present a new part-of-speech tagger that demonstrates the following ideas: (i) explicit use of both preceding and following tag contexts via a dependency network representation, (ii) broad use of lexical features, including jointly conditioning on multiple consecutive words, (iii) effective use of priors in conditional loglinear models, and (iv) fine-grained modeling of unknown word features. Using these ideas together, the resulting tagger gives a 97.24% accuracy on the Penn Treebank WSJ, an error reduction of 4.4% on the best previous single automatically learned tagging result.
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.
Two apparently opposing DOP models exist in the literature: one which computes the parse tree involving the most frequent subtrees from a treebank and one which computes the parse tree involving the fewest subtrees from a treebank. This paper proposes an integration of the two models which outperforms each of them separately. Together with a PCFG-reduction of DOP we obtain improved accuracy and efficiency on the Wall Street Journal treebank. Our results show an 11% relative reduction in error rate over previous models, and an average processing time of 3.6 seconds per WSJ sentence.
The purpose of this study was to investigate preservice teacher intensity, time on and off task, and effectiveness in relation to in-service teacher retention/attrition. Senior music education majors (N = 150) made a videotape of their “best” teaching during the final weeks of student teaching. Excerpts were observed, evaluated, and teacher time on/off task was recorded using the Simple Recording Interface for Behavioral Observation (Duke & Farra, 1996). Each subject's teaching status as K—college music educator in 1995 and 2001 was then determined. No differences were found for intensity, effectiveness, or time on / off task in relation to teaching status assessed, suggesting that preservice teaching performance does not indicate continuing participation in music education as an in-service music teacher. Further analysis revealed that mean interval duration of teacher off task affects ratings of intensity and effectiveness more than do frequency and duration of on task. Application to teacher training and suggestions for further areas of research are discussed.
This paper deals with wordnet development tools. It presents a designed and developed system for lexical database editing, which is currently employed in many national wordnet building projects. We discuss basic features of the tool as well as more elaborate functions that facilitate linguistic work in multilingual environment.
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.
OBJECTIVES: While several sources of biomedical knowledge are available, these resources are often highly specialized and usually not suitable for a lay audience. This paper evaluates whether concepts needed for molecular biology and genetic diseases are present in WordNet, the electronic lexical database. METHODS: Terms for four broad categories of concepts (phenotype, molecular function, biological process, and cellular component) were extracted from LocusLink and mapped to WordNet. All terms from the Gene Ontology database (gene products and ontology concepts) were also mapped to WordNet in order to evaluate its global coverage of the domain. Additionally, we tested two methods for improving the mapping of genetic disease names to WordNet. RESULTS: The coverage of concepts ranged from 0% (gene product symbols) to 2.8% (cellular components). Removing specialization markers from the terms and using synonyms significantly increased the rate of mapping of genetic disease names to WordNet. CONCLUSIONS: Many of the most common single gene disorders are present in WordNet, as well as many high-level concepts in Gene Ontology. Therefore, WordNet is likely to be a useful source of lay knowledge in the framework of a consumer health information system on genetic diseases.
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.
The paper reports on the design and construction of a multi-layered corpus of Italian, annotated at the syntactic and lexico-semantic levels, whose development is supported by dedicated software augmented with an intelligent interface. The issue of evaluating this type of resource is also addressed.
The thesis is titled Mario Monteleone Electronic Dictionaries and Lexicography. Uses language with lexical databases provided concerning the relationship (and potential) between traditional lexicography and computational linguistics. In particular, Mr. Monteleone is trying to establish whether some of the lexicography analytical limits can be exceeded through the application of research methods in computational linguistics. To this end, Mr. Monteleone is conducting a detailed analysis of the two disciplines, of their traditional instruments, ie dictionaries paper and electronic dictionaries, in terms of their construction methods and their application areas. The result is a handbook for traditional lexicographers and computer focused on the identification and classification of linguistic data for the accomplishments of paper and electronic dictionaries.
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.
▪ 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.
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.
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
BACKGROUND: Human affective responses appear to be regulated by limbic and paralimbic circuits. However, much less is known about the neurochemical systems engaged in this regulation. The mu-opioid neurotransmitter system is distributed in, and thought to regulate the function of, brain regions centrally implicated in affective processing. OBJECTIVE: To examine the involvement of mu-opioid neurotransmission in the regulation of affective states in healthy human volunteers. DESIGN: Measures of mu-opioid receptor availability in vivo were obtained with positron emission tomography and the mu-opioid receptor selective radiotracer [11C]carfentanil during a neutral state and during a sustained sadness state. Subtraction analyses of the binding potential maps were then performed within subjects, between conditions, on a voxel-by-voxel basis. SETTING: Imaging center at a university medical center. PARTICIPANTS: Fourteen healthy female volunteers. Intervention Sustained neutral and sadness states, randomized and counterbalanced in order, elicited by the cued recall of an autobiographical event associated with that emotion. MAIN OUTCOME MEASURES: Changes in mu-opioid receptor availability and negative and positive affect ratings between conditions. Increases or reductions in the in vivo receptor measure reflect deactivation or activation of neurotransmitter release, respectively. RESULTS: The sustained sadness condition was associated with a statistically significant deactivation in mu-opioid neurotransmission in the rostral anterior cingulate, ventral pallidum, amygdala, and inferior temporal cortex. This deactivation was reflected by increases in mu-opioid receptor availability in vivo. The deactivation of mu-opioid neurotransmission in the rostral anterior cingulate, ventral pallidum, and amygdala was correlated with the increases in negative affect ratings and the reductions in positive affect ratings during the sustained sadness state. CONCLUSIONS: These data demonstrate dynamic changes in mu-opioid neurotransmission in response to an experimentally induced negative affective state. The direction and localization of these responses confirms the role of the mu-opioid receptor system in the physiological regulation of affective experiences in humans.
Crosslinguistically vocatives are an underexplored linguistic phenomenon and in different languages they can be highly idiosyncratic and complex (Levinson, 1987, p.71). Therefore, the problem, which is discussed in this paper, is not a language-specific one, in spite of the fact that most of the languages have their own repositories for marking the role of the addressee in the communicative utterances. In our opinion this linguistic phenomenon needs its adequate treatment in HPSG because of three main reasons: 
 
 The vocative is supposed to be present on two levels: syntax and pragmatics. Therefore it needs more elaborate interpretation on the interface side, which, in HPSG, is more developed for morphology/syntax and syntax/semantics than syntax/pragmatics. Note that a challenge for the theory is the semantic weight of the vocatives with respect to the head sentence. 
 It will be useful for HPSG-oriented implementations, especially treebanks and dialogue systems. 
 On prosodic grounds the vocatives are often viewed as being 'side or extended parts' of the sentence and therefore - very close to the parenthetical constructions. From our point of view, both phenomena are pragmatic and hence, the treatment of vocative, presented here, could be generalized to cover other phenomena of pragmatic nature. 
 
 In our work the vocatives are viewed through the possibility of the integration/separation of their pragmatic, syntactic and semantic properties.
Abstract Two experiments examined the effects of word familiarity on word recognition and text comprehension during silent reading. Readers' eye movements were monitored as they read sentences containing words that varied in familiarity as assessed by printed estimates of word frequency, subjective ratings of familiarity, and a multiple‐choice test of meaning knowledge. Effects of word frequency were unaffected by differences in subjective familiarity rating for high frequency words. Differential effects of familiarity rating were observed in low frequency conditions. In addition, processing time on high and low frequency words did not differ when familiarity was held constant for moderately familiar words. Readers spent more initial processing time on novel words than familiar words. Performance on a vocabulary test administered after the reading session demonstrated that readers successfully acquired and retained new word meanings. Finally, reanalysis of word processing time as a function of vocabulary test performance demonstrated a systematic relationship between online processing patterns and memory for novel word meaning.
Pleasant stimuli typically elicit greater electromyographic (EMG) activity over zygomaticus major and less activity over corrugator supercilii than do unpleasant stimuli. To provide a systematic comparison of these 2 measures, the authors examined the relative form and strength of affective influences on activity over zygomaticus major and corrugator supercilii. Self-reported positive and negative affective reactions and facial EMG were collected as women (n = 68) were exposed to series of affective pictures, sounds, and words. Consistent with speculations based on known properties of the neurophysiology of the facial musculature, results revealed a stronger linear effect of valence on activity over corrugator supercilii versus zygomaticus major. In addition, positive and negative affect ratings indicated that positive and negative affect have reciprocal effects on activity over corrugator supercilii, but not zygomaticus major.
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 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.
This paper discusses an automatic, data-driven approach to treebank error detection. The approach adapts the use of so-called variation n-grams as defined in Dickinson and Meurers (2003) for the detection of inconsistent part-of-speech annotations to syntactic annotation. The underlying idea is to define a consistency test for the mapping from recurring strings to their syntactic annotation. The paper illustrates with a case study based on the WSJ treebank that the method successfully detects inconsistencies in syntactic category annotation. Since such inconsistencies are typically introduced by humans, our method works best for large corpora that have been annotated manually or semi-automatically, which is generally the case for current syntactic and other high-level annotation. \n \nOur work serves two main purposes for treebank improvement. It is a means for finding erroneous variation in a corpus, which can then be corrected. And it provides feedback for the development of empirically adequate standards for syntactic annotation, showing which distinctions are difficult to maintain over an entire corpus. Additionally, as a method for comparing syntactic annotation, our work could have uses for interannotator agreement testing and parser evaluation.
We investigate the performance of the Structured Language Model (SLM) in terms of perplexity (PPL) when its components are modeled by connectionist models. The connectionist models use a distributed representation of the items in the history and make much better use of contexts than currently used interpolated or back-off models, not only because of the inherent capability of the connectionist model in fighting the data sparseness problem, but also because of the sublinear growth in the model size when the context length is increased. The connectionist models can be further trained by an EM procedure, similar to the previously used procedure for training the SLM. Our experiments show that the connectionist models can significantly improve the PPL over the interpolated and back-off models on the UPENN Treebank corpora, after interpolating with a baseline trigram language model. The EM training procedure can improve the connectionist models further, by using hidden events obtained by the SLM parser.
This article reports the outcome of a publicly funded research project titled "Redesign of the British Sign Language (BSL) Notation System with a New Font for Use in ICT," which ran from September 2000 to December 2001. The aim of the project was to redesign the British Sign Language variant of Stokoe notation (as used in the BSL/English Dictionary) for practical use in information technology systems and software, such as lexical databases, word-processing packages, and teaching and learning applications. The project�s objectives tackled design issues, not sign linguistic system-level problems. The project resulted in two new type designs for writing BSL Stokoe, united into a larger type family called William C., in memory of William C. Stokoe. I anticipate that the new type design proposals will make the provision of searchable BSL Stokoe notation in database designs a possible next step in lexicographic and other sign linguistic research projects.
We demonstrate that an unlexicalized PCFG can parse much more accurately than previously shown, by making use of simple, linguistically motivated state splits, which break down false independence assumptions latent in a vanilla treebank grammar. Indeed, its performance of 86.36% (LP/LR F1) is better than that of early lexicalized PCFG models, and surprisingly close to the current state-of-the-art. This result has potential uses beyond establishing a strong lower bound on the maximum possible accuracy of unlexicalized models: an unlexicalized PCFG is much more compact, easier to replicate, and easier to interpret than more complex lexical models, and the parsing algorithms are simpler, more widely understood, of lower asymptotic complexity, and easier to optimize.
The paper reports on the design and construction of a multilayered corpus of Italian, annotated at the syntactic and lexico-semantic levels, whose development is supported by dedicated software augmented with an intelligent interface. The issue of evaluating this type of resource is also addressed.
We present a neural-network-based statistical parser, trained and tested on the Penn Treebank. The neural network is used to estimate the parameters of a generative model of left-corner parsing, and these parameters are used to search for the most probable parse. The parser's performance (88.8% F-measure) is within 1% of the best current parsers for this task, despite using a small vocabulary size (512 inputs). Crucial to this success is the neural network architecture's ability to induce a finite representation of the unbounded parse history, and the biasing of this induction in a linguistically appropriate way.
The 200 million word corpus of the Bank of English was annotated morphologically and syntactically using the English Constraint Grammar analyser, a rulebased shallow parser developed at the Research Unit for Computational Linguistics, University of Helsinki. We discuss the annotation system and methods used in the corpus work, as well as the theoretical assumptions of the Constraint Grammar syntax. Based on our experience in large-scale corpus work, we argue for a deeper and more explicit, dependency-based syntactic representation. We present a new practical parsing system, the Functional Dependency Grammar parser, developed from the Constraint Grammar system, and discuss its suitability for treebank annotation.
This report details experimental results of using stochastic disambiguation models for parsing sentences from the Redwoods treebank (Oepen et al., 2002). The goals of this paper are two-fold: (i) to report accuracy results on the more highly ambiguous latest version of the treebank, as compared to already published results achieved by the same stochastic models on a previous version of the corpus, and (ii) to present some newly developed models using features from the HPSG signs, as well as the MRS dependency graphs.
In this paper, we propose a method for analyzing word-word dependencies using deterministic bottom-up manner using Support Vector machines. We experimented with dependency trees converted from Penn treebank data, and achieved over 90 % accuracy of word-word dependency. Though the result is little worse than the most up-to-date phrase structure based parsers, it looks satisfactorily accurate considering that our parser uses no information from phrase structures. 1
The linguistic annotation of natural language corpora is one of the main areas of computational linguistics. Much energy has been devoted to building large syntactically annotated corpora, which are also called treebanks after the phrase-structure trees they contain. For some time now, functional information, i.e. information on whether a constituent functions as e. g. subject, object or adverbial, has also been included in the annotation. Yet, while at first glance this may not seem to be a venture too complicated, matters are not always as easy as they seem.
OBJECTIVE: To compare the subjective image quality of the newer generation Schick CDR detector employing complementary metal oxide semiconductor (CMOS) technology with images using the earlier generation charge-coupled device (CCD) Schick CDR detector. METHODS: All radiographic images were made using the same formalin-fixed adult cadaver maxilla with surrounding natural soft tissues in place. The X-ray generator used was a Villa Sistemi Medicali Diamatic srl AP/Explor X operated at 70 kVp and 8 mA. The source-to-detector distance was set at 38 cm and an optical bench was used to ensure reproducible beam geometry. A range of exposures was applied for both detectors. A panel of nine dentists independently observed and evaluated images made at each exposure. For both detectors, the three images ranked highest were randomized for re-evaluation in panels of six images. Each image was repeated randomly a total of 10 times. Features chosen as observation points were: (1) proximal dental caries; (2) gingival soft tissues; (3) cortical bone; (4) root canal space; (5) root apices; (6) periodontal ligament space; and (7) endodontic instrument tip clarity. Comparisons were made by use of odds ratio analysis applying a 95% confidence level. Interrater and intrarater reliabilities were computed to assess consistency in observer ratings. RESULTS: The CMOS sensor was rated as outperforming its CCD predecessor for depiction of cortical bone and root apices; the CCD detector was only rated superior for depiction of root canal space. No significant difference was found between the two detectors in perceived depiction of proximal dental caries, gingival soft tissues, periodontal ligament space or endodontic instruments. Combining rating scores from each of the tasks, CMOS and CCD detectors had a similar proportion of image ratings of excellent, acceptable and poor. CONCLUSIONS: Regarding subjective image quality, the Schick CMOS and CCD detectors were perceived to produce radiographic images of similar overall quality.