Papers reviewed and determined not to be word norm studies. Use the flag icon to report errors or suggest re-inclusion.
18625 papers
In the last decade, given the availability of corpora in several distinct languages, research on multilingual part-of-speech tagging started to grow. Amongst the novelties there is mWANN-Tagger (multilingual weightless artificial neural network tagger), a weightless neural part-of-speech tagger capable of being used for mostly-suffix-oriented languages. The tagger was subjected to corpora in eight languages of quite distinct natures and had a remarkable accuracy with very low sample deviation in every one of them, indicating the robustness of weightless neural systems for part-of-speech tagging tasks. However, mWANN-Tagger needed to be tuned for every new corpus, since each one required a different parameter configuration. For mWANN-Tagger to be truly multilingual, it should be usable for any new language with no need of parameter tuning. This article proposes a study that aims to find a relation between the lexical diversity of a language and the parameter configuration that would produce the best performing mWANN-Tagger instance. Preliminary analyses suggested that a single parameter configuration may be applied to the eight aforementioned languages. The mWANN-Tagger instance produced by this configuration was as accurate as the language-dependent ones obtained through tuning. Afterwards, the weightless neural tagger was further subjected to new corpora in languages that range from very isolating to polysynthetic ones. The best performing instances of mWANN-Tagger are again the ones produced by the universal parameter configuration. Hence, mWANN-Tagger can be applied to new corpora with no need of parameter tuning, making it a universal multilingual part-of-speech tagger. Further experiments with Universal Dependencies treebanks reveal that mWANN-Tagger may be extended and that it has potential to outperform most state-of-the-art part-of-speech taggers if better word representations are provided.
In statistical parsing, the probabilistic models are used to evaluate the possibility of each candidate parse tree, where the parse tree with the largest probability is deemed to be the final result of the parsing. Therefore, the core of statistical parsing is a probabilistic evaluation model. The main difference among the various probabilistic evaluation models lies in which types of features in the context are used to assign the probabilities to the parse trees. Various probabilistic evaluation models have been proposed in the field of statistical parsing, where different models use different feature types. How to evaluate a feature type's predictive power for the parsing tree? The paper proposes an information theory based feature type analysis model. Using the method, we can quantitatively analyze the power of different feature types for syntactic structure prediction from the viewpoint of information theory. The basic idea is that we use entropy and conditional entropy to measure whether a feature type grasps some of the information for syntactic structure prediction. If the average uncertainty of the syntactic structures declines apparently, the feature type is deemed to have grasped some intrinsic linguistic information in the context that has close relation to the syntactic structure. Using Penn Treebank as training and testing set, our experiment quantitatively analyze the different feature types' predictive power for syntactic structure predictive power for syntactic structure prediction in a systematic way and draws a series of conclusions which reflect the predictive power of different feature types and feature type combination for syntactic parsing.
Natural language processing technologies offer ease-of-use of computers for average users, and ease-of-access to on-line information. Natural language, however, is complex, and the traditional methods of parsing with a single grammar and parser may result in an inefficient and large system that is difficult to maintain and is fragile in dealing with language irregularities. This paper begins by reviewing an alternative effort in grammar decomposition (also known as grammar partitioning) for natural language parsing, which aims to alleviate these problems. We then propose a novel automatic approach for grammar partitioning, in comparison with a random method of partitioning. Our experiments show that syntactic GLR parsing is formidable for the Wall Street Journal corpus in the Penn Treebank when a single grammar is used. This is due to too many grammar rules for parsing table generation. However, grammar partitioning solves the problem and offers a viable alternative. Our results also show that our automatic grammar partitioning method based on the mutual information criterion fares better than a random partitioning method and exhibits efficiency in parsing as well as high parse coverage. 1
Low interrater reliability coefficients are a common problem for behavior rating scales. One hypothesis to account for this is that raters have different frames of reference from which to judge behaviors. In the present study, the interrater reliability of the Devereux Behavior Rating Scale-School Form was examined, and the hypothesis that teacher frame of reference influences ratings was explored. Special and general education teachers rated the behavior of 51 children with emotional disturbance (ED), and general education teachers inde pendently rated the behavior of 51 matched control children. Interrater reliability coeffi cients were higher for the general education sample than for the sample of children with ED. Limited support was found for the hypoth esis that frame of reference may affect ratings. Findings suggest that many factors influence ratings and that teachers may benefit from rater training.
The form and content of the lexical database of Auslan (Australian Sign Language) is described and explained. The type of database utilized and its precise structure (relational or flat, the type and number of fields, the design of the data entry interface, etc.) is first described. This is followed by a detailed description of the types of information registered in the database: phonological, definitional, bilingual (English-based glossing), grammatical, and semantic. The non-gloss based representations of each sign record (graphic, video, and transcription) that are used in the lexical database are then discussed. Finally, the compatibility of the Auslan lexical database with other lexical databases is examined. The paper concludes with a discussion of the possibility of building an extensive “universal” database of signs that could centralize lexical information from scores of signed languages and facilitate cross-linguistic investigations of lexis and phonology.
Computational learning of natural language is often attempted without using the knowledge available from other research areas such as psychology and linguistics. This can lead to systems that solve problems that are neither theoretically or practically useful. In this paper we present a system CLL which aims to learn natural language syntax in a way that is both computationally effective and psychologically plausible. This theoretically plausible system can also perform the practically useful task of unsupervised learning of syntax. CLL has then been applied to a corpus of declarative sentences from the Penn Treebank (Marcus et al., 1993; Marcus et al., 1994) on which it has been shown to perform comparatively well with respect to much less psychologically plausible systems, which are significantly more supervised and are applied to somewhat simpler problems.
This paper describes a framework for examining the effects of the cognitive complexity of tasks on language production and learner perceptions of task difficulty, and for motivating sequencing decisions in task-based syllabuses. Results of a study of the relationship between task complexity, difficulty, and production show that increasing the cognitive complexity of a direction-giving map task significantly affects speaker-information-giver production (more lexical variety on a complex version and greater fluency on a simple version) and hearer-information-receiver interaction (more confirmation checks on a complex version). Cognitive complexity also significantly affects learner perceptions of difficulty (e.g. a complex version is rated significantly more stressful than a simple version). Task role significantly affects ratings of difficulty, though task sequencing (simple to complex versus the reverse sequence) does not. However, sequencing does affect the accuracy and fluency of speaker production. Implications of the findings for task-based syllabus design and further research into task complexity, difficulty, and production interactions are discussed.
Abstract This study examines judgments of male self-control regarding sexual aggression in dating situations. A survey was conducted using vignettes that described situations where a hypothetical man attempts to have intercourse with his female date, she resists, and he considers various ways of overcoming her resistance. Survey respondents judged these vignettes. The goals of the study were to estimate the prevalence of beliefs about low male self-control, and to examine how contextual information affects these assessments. Contrary to expectations derived from the literature, the large majority of respondents attributed high levels of self-control to men across a wide range of scenarios. As expected, the man's alcohol intoxication reduced perceived control. Against expectations, consensual foreplay and prior sex did not affect ratings. Implications for social learning and rational choice theories of sexual aggression are discussed, with particular emphasis on Ajzen's (1988) theory of planned behavior.
Cross-language information retrieval (CLIR), where queriesand documents are in different languages, has of late become one ofthe major topics within the information retrieval community. Thispaper proposes a Japanese/English CLIR system, where we combine aquery translation and retrieval modules. We currently target theretrieval of technical documents, and therefore the performance of oursystem is highly dependent on the quality of the translation oftechnical terms. However, the technical term translation is stillproblematic in that technical terms are often compound words, and thusnew terms are progressively created by combining existing basewords. In addition, Japanese often represents loanwords based on itsspecial phonogram. Consequently, existing dictionaries find itdifficult to achieve sufficient coverage. To counter the firstproblem, we produce a Japanese/English dictionary for base words, andtranslate compound words on a word-by-word basis. We also use aprobabilistic method to resolve translation ambiguity. For the secondproblem, we use a transliteration method, which corresponds wordsunlisted in the base word dictionary to their phonetic equivalents inthe target language. We evaluate our system using a test collectionfor CLIR, and show that both the compound word translation andtransliteration methods improve the system performance.
This study provides the information of reliability and implications of narrative essay tests for measuring achievement motive as a part of employee selection processes. To develop the key achievement motive explicit and concrete rating criterion, the TAT scoring method was applied to data (n=100) of essay tests gathered from seven raters. Three raters were chosen from entry level workers and the other four were professional writers of verbal testing items, forming two contrast groups. The reliability of the achievement motive ratings was calculated for each group by the interrater reliability approach, resulting that there was no significant difference between the groups. Coefficients of each group's achievement motive, general mental ability and personality traits were calculated, suggesting the possibility that the achievement motive ratings thus derived from essay tests implies individuality.
Chunk parsing has focused on the recognition of partial constituent structures at the level of individual chunks. Little attention has been paid to the question of how such partial analyses can be combined into larger structures for complete utterances.The TüSBL parser extends current chunk parsing techniques by a tree-construction component that extends partial chunk parses to complete tree structures including recursive phrase structure as well as function-argument structure. TüSBL's tree construction algorithm relies on techniques from memory-based learning that allow similarity-based classification of a given input structure relative to a pre-stored set of tree instances from a fully annotated treebank.A quantitative evaluation of TüSBL has been conducted using a semi-automatically constructed treebank of German that consists of appr. 67,000 fully annotated sentences. The basic PARSEVAL measures were used although they were developed for parsers that have as their main goal a complete analysis that spans the entire input. This runs counter to the basic philosophy underlying TüSBL, which has as its main goal robustness of partially analyzed structures.
Grammars are valuable resources for natural language processing. We divide the process of grammar development into three tasks: selecting a formalism, defining the prototypes, and building a grammar for a particular human language. After a brief discussion about the first two tasks, we focus on the third task. Traditionally, grammars are built by hand and there are many problems with this approach. To address these problems, we built two systems that automatically generate grammars. The first system (LexOrg) solves two major problems in grammar development: namely, the redundancy caused by the reuse of structures in a grammar and the lack of explicit generalizations over the structures in a grammar. LexOrg takes several types of specification as input and combines them to automatically generate a grammar. The second system (LexTract) extracts Lexicalized Tree Adjoining Grammars (LTAGs) and Context-free Grammars (CFGs) from Treebanks, and builds derivation trees that can be used to train statistical LTAG parsers directly. In addition to creating Treebank grammars and producing training materials for parsers, LeXTract is also used to evaluate the coverage of existing hand-crafted grammars, to compare grammars for different languages, to detect annotation errors in Treebanks, and to test certain linguistic hypotheses. LexOrg and LeXTract provide two different perspectives on grammars. In LexOrg, elementary trees in an LTAG grammar are the result of combining language specifications such as tree descriptions. In LeXTract, elementary trees are building blocks of syntactic structures in a Treebank. LexOrg makes explicit the language specifications that form elementary trees, whereas LeXTract makes explicit the elementary trees that form syntactic structures. The systems provide a rich set of tools for language description and comparison that greatly enhances our ability to build and maintain grammars and Treebanks effectively.
Although generic (i.e. domain independent) and specialized (i.e. domain specific) lexical resources are usually developed with different aims, an integrated consultation seems to be necessary for many NLP based applications. We describe an integration procedure based on the definition of plug-in relations that are established to manage overlaps and inconsistencies between the two resources. The approach has been experimented connecting ItalWordNet, a generic lexical database for Italian, and Economic-WordNet, a specialized wordnet for the economic and financial domain. In this paper we address the issue of integrating the information included in a generic lexical database with the information included in a specialized (i.e. domain specific) lexical database. We restrict our investigation to wordnet-like lexical resources, that is lexical databases whose model is derived from WordNet (Miller, 1990) (Fellbaum, 1998). The aim is to define a set of procedures to allow an of the two resources, such that overlapping senses are merged and conflicting situations are properly managed. Our starting point are two existing wordnets for Italian: ItalWordnet (IWN)1, a generic wordnet deriving from EuroWordNet (Vossen, 1999), and Economic-wordnet (ECOWN), a wordnet for the economic and financial domain. The current application scenario is a recommender system (Magnini and Strapparava, 2001), which includes a 1 This research has been supported by SI-TAL (Integrated System for the Automatic Treatment of Language), a National Project devoted to the creation of large linguistic resources and software for Italian written and spoken language processing. document processing module based on a light form of word sense disambiguation. The experimental domain is that of financial news. As a first attempt to use the two resources in conjunction, a “specific-first” strategy was implemented, which, given a lemma in the document, first looks up in the specific wordnet and just in case of failure resorts to the generic wordnet. An evident drawback of this approach is that it does not cover cases of words belonging to both of the databases when the correct interpretation of the word is placed in the generic wordnet. For instance, in the phrase “building its share market” (taken from a financial news), the word “share” is used with the generic meaning of “portion”, while in the specialized wordnet we would have the economic sense of “share” as part of the capital stock. To overcome this limit we tried with a “union” strategy which, for a given lemma, considers the sum of the senses for that lemma in the two resources. The problem here is that senses for words belonging to both the databases tend to proliferate, making disambiguation harder. What seems necessary is a deeper integration of the respective word senses, such that overlapping senses are merged and conflicting situations are detected and solved. Our scenario allows some significant simplifications with respect to the general problem of merging two distinct ontologies (Hovy, 1998). On the one side we have a specialized database, whose content is supposed to be more accurate and precise as far as specialized information is concerned; on the other side we can assume that the generic resource guarantees a more uniform coverage as far as high level senses are concerned. These two assumptions provide us with a powerful precedence criterion to be used for managing inheritance in the integration procedure. The contribution of our work consists in a!# $ % &(' ) approach which allows the connecting of the generic and the specialized wordnets in a flexible and modular way. This is realized by means of a semi-automatic procedure with four main steps: (i) first, a minimum set of specialized “basic synsets” is identified; (ii) basic synsets are aligned to corresponding generic synsets and a particular plug-in configuration is selected; (iii) for each plug-in configuration a merging algorithm reconstructs the corresponding portion of the integrated wordnet; (iv) possible inconsistencies are solved. There are two main benefits of this approach. First, already existing specialized resources can be connected to a generic resource without any change in the resource being necessary, a part from the data conversion into a wordnet-like format. Second, the inheritance of linguistic oriented information makes the specialized resources usable in existing wordnet-based applications. The paper is structured as follows. Section 2 presents the lexical resources we have used. Section 3 introduces the basic notions of the plug-in approach with some technical details. Section 4 describes the plug-in procedure and reports the results of an application of the approach. Section 5 places our proposal in the context of related works. * +-,., / 0 132.5476 8,91 0 2;:, 4@?BA / 45., C D In this work we assume a taxonomic wordnetlike structure of the lexicon, where nodes in the hierarchy are synsets (i.e. synonym sets), and a rather large set of conceptual relations (e.g. Part-of, Cause, Hypernymy, Pertains-to, etc.) are available to build a semantic net among synsets. The EuroWordNet model has been adopted, which is rich enough to encompass most of the relations used in existing non wordnet-like terminological databases. We focus on the integration of already existing generic and specialized wordnets; both the data acquisition modalities and the evaluation of the quality of the resources do not affect our approach. As for generic lexical databases, typically they contain knowledge with no specific coverage and much attention is placed on coding linguistic-oriented information, such as subcategorization relations and fine-grained sense distinctions. There are several examples of existing generic lexical databases, including the English WordNet (Miller, 1990), monolingual wordnets for several European languages (e.g. Dutch, Spanish, Italian, Basque, etc.), the SENSUS (Hovy, 1998) and the Mikrokosmos (Mahesh, 1996) ontologies. Specialized databases focus on a certain domain, providing sub-hierarchies of highly specialized concepts with a limited use of lexical and linguistic relations. Synset variants tend to assume the shape of complex terms (i.e. multiwords) and the role of the domain expert is crucial for establishing correct relations. In addition high level knowlegde (i.e. the top ontology) tend to be simplified and domain oriented. Many specialized lexical databases have been developed, particularly for concrete applications, including, for example, a taxonomy for the medical domain (Gangemi et E F., 1999), the Art and Architecture Getty Thesaurus and the Getty Thesaurus of Geographical Names. The plug-in model we present has been applied within the SI-TAL project to connect a generic wordnet and a specialized wordnet that have been created independently. GIH J K LNM O P Q R H (Roventini et S T., 2000) created as part of the EuroWordNet project and further developed through the introduction of adjectives and adverbs, is the lexical database involved in the plug-in as a generic resource and consists of about 45,000 lemmas. U5V W X W Y@Z V []\^W _ ` acb d is a specialized wordnet for the economic domain and consists of about 5,000 lemmas distributed in about 4,700 synsets. Table 1 summarizes the quantitative data of the two resources considered.
The form and content of the lexical database of Auslan (Australian Sign Language) is described and explained. The type of database utilized and its precise structure (relational or flat, the type and number of fields, the design of the data entry interface, etc.) is first described. This is followed by a detailed description of the types of information registered in the database: phonological, definitional, bilingual (English-based glossing), grammatical, and semantic. The non-gloss based representations of each sign record (graphic, video, and transcription) that are used in the lexical database are then discussed. Finally, the compatibility of the Auslan lexical database with other lexical databases is examined. The paper concludes with a discussion of the possibility of building an extensive “universal” database of signs that could centralize lexical information from scores of signed languages and facilitate cross-linguistic investigations of lexis and phonology.
Abstract. We present a suite of applications used for the Italian Treebank which share their linguistic processor and end up finally in higher level annotation tool called “FILES”. The first application “FILES ” – Fully Integrated Linguistic Environment for Syntactic and Functional Annotation- is a prototype for a fully integrated linguistic environment for syntactic functional annotation of corpora. It takes as input tagged and disambiguated tokenized texts, a file containing the same text split into sentences, and a files containing the morphosyntactic and semantic features associated to each tagged token. Tokens may be aither single words, polywords, abbreviations or punctuation marks. An as yet separated module of “FILES ” is the syntactic constituency annotation environment which uses a shallow parser on the same tagged files and produces a fully bracketed output where each sentence is a record. Files contaning bracketed sentences are given as input to Syntactic Constituency Annotation Tool “SCAT” for manual verification. 1
In an attempt to establish a possible ‘norm’ for the distribution of translation modalities in English → Portuguese translational relationship, a varied sample of three different text typologies (legal, technical, and corporate) with six representative texts of each typology was compared, producing a total of 9,000 lexical items. By applying Vinay & Darbelnet’s and Aubert’s models, it was possible to obtain a basic pattern of the distribution of the most used translation modalities as well as to verify certain variables, such as the correlation between a higher or lower fluctuation in the frequency of the modalities and different types of texts. From three different levels of data analysis, we observed a translation hierarchy in relation to the three most frequent categories: literal translation, transposition and modulation and also that legal texts on the one hand, and technical and corporate texts on the other, seemed to organise themselves into two major groups. We also obtained some elements which would enable us to sketch a correlation between the modalities of literal translation and technical and corporate texts, as well as a correlation between the modalities of modulation and transposition with modulation and legal texts.
We present a stochastic parsing system consisting of a Lexical-Functional Grammar (LFG), a constraint-based parser and a stochastic disambiguation model. We report on the results of applying this system to parsing the UPenn Wall Street Journal (WSJ) treebank. The model combines full and partial parsing techniques to reach full grammar coverage on unseen data. The treebank annotations are used to provide partially labeled data for discriminative statistical estimation using exponential models. Disambiguation performance is evaluated by measuring matches of predicate-argument relations on two distinct test sets. On a gold standard of manually annotated f-structures for a subset of the WSJ treebank, this evaluation reaches 79% F-score. An evaluation on a gold standard of dependency relations for Brown corpus data achieves 76% F-score.
This paper begins with the significance of language information processing technology for the progress of an information society, and then introduces that Institute of Computational Linguistics (ICL) of Peking University and other language resource data on web sites, such as HowNet, WordNet, TreeBank., achieved the results of these researches and actualities of the language resources at the foundational researches on language information processing. The last section of this paper prints out the tasks for further studies.
Chunk parsing has focused on the recognition of partial constituent structures at the level of individual chunks. Little attention has been paid to the question of how such partial analyses can be combined into larger structures for complete utterances. Such larger structures are not only desirable for a deeper syntactic analysis. They also constitute a necessary prerequisite for assigning function-argument structure.The present paper offers a similarity-based algorithm for assigning functional labels such as subject, object, head, complement, etc. to complete syntactic structures on the basis of prechunked input.The evaluation of the algorithm has concentrated on measuring the quality of functional labels. It was performed on a German and an English treebank using two different annotation schemes at the level of function-argument structure. The results of 89.73 % correct functional labels for German and 90.40 % for English validate the general approach.
Considering the difficulties inherent in the manual construction of natural language parsers, we have designed and implemented our system GRIND which is capable of learning a sequence of context-dependent parsing actions from an arbitrary corpus containing labelled parse trees. Being trained and tested on corpus SUSANNE, GRIND reaches the accuracy of 96 % and the recall of 68 %.
CL Research's word-sense disambiguation (WSD) system is part of the DIMAP dictionary software, designed to use any full dictionary as the basis for unsupervised disambiguation. Official SENSEV AL-2 results were generated using WordNet, and separately using the New Oxford Dictionary of English (NODE). The disambiguation functionality exploits whatever information is made available by the lexical database. Special routines examined multiword units and contextual clues (both collocations, definition and example content words, and subject matter analyses); syntactic constraints have not yet been employed. The official coarsegrained precision was 0.367 for the lexical sample task and 0.460 for the all-words task (these are actually recall, with actual precision of 0.390 and 0.506 for the two tasks). NODE definitions were automatically mapped into WordNet, with precision of0.405 and 0.418 on 75 % and 70 % mapping for the lexical sample and all-words tasks, respectively, comparable to WordNet. Bug fixes and implementation of incomplete routines have increased the precision for the lexical sample to 0.429 (with many improvements still likely).
The current state of affairs is characterised as one in which general SLA models have syntax as their core and pay less and variable attention to other linguistic levels, notably lexis. In order to improve the current situation we need involvement from both the vocabulary research community and SLA model builders. It is demonstrated how the former group readily borrows key concepts from psycholinguistics and SLA theory and rethinks them from a lexical point of view. However, such borrowing and recasting is often done in a piecemeal fashion to fit specific research issues. As for SLA model builders, some examples are discussed that are regarded as serious attempts at integrating lexis into a particular acquisition model. One is L2 reading research and vocabulary acquisition through reading, which illustrates a high degree of integration with common research goals and mutual theoretical inspiration. A second example underlines the fact that there is an obvious potential for including lexis in the ‘focus on form’ movement. It is our contention that more attention to lexis should supplement the predominantly grammatical ‘focus on form’ that is the current norm.
In this paper we discuss the need for corpora with a variety of annotations to provide suitable resources to evaluate different Natural Language Processing systems and to compare them. A supervised machine learning technique is presented for translating corpora between syntactic formalisms and is applied to the task of translating the Penn Treebank annotation into a Categorial Grammar annotation. It is compared with a current alternative approach and results indicate annotation of broader coverage using a more compact grammar.
In this paper a new similarity-based learning algorithm, inspired by string edit-distance (Wagner and Fischer, 1974), is applied to the problem of bootstrapping structure from scratch. The algorithm takes a corpus of unannotated sentences as input and returns a corpus of bracketed sentences. The method works on pairs of unstructured sentences or sentences partially bracketed by the algorithm that have one or more words in common. It finds parts of sentences that are interchangeable (i.e. the parts of the sentences that are different in both sentences). These parts are taken as possible constituents of the same type. While this corresponds to the basic bootstrapping step of the algorithm, further structure may be learned from comparison with other (similar) sentences. We used this method for bootstrapping structure from the flat sentences of the Penn Treebank ATIS corpus, and compared the resulting structured sentences to the structured sentences in the ATIS corpus. Similarly, the algorithm was tested on the OVIS corpus. We obtained 86.04 % non-crossing brackets precision on the ATIS corpus and 89.39 % non-crossing brackets precision on the OVIS corpus.
Hemispheric Effects of Concreteness in Pictures and Words Daniel J. Casasanto 1 (dcasasan@mail.med.upenn.edu) John Kounios 2 (jkounios@cattell.psych.upenn.edu) John A. Detre 1 (detre@mail.med.upenn.edu) Department of Neurology 1 and Psychology 2, University of Pennsylvania 3400 Spruce Street, Philadelphia, PA 19104 USA Introduction Functional Magnetic Resonance Imaging (fMRI) studies have demonstrated differently lateralized activation during episodic memory encoding for verbal and pictorial stimuli, as predicted by the material-specific model (1). Results are interpreted as consistent with Dual Coding Theory, which posits parallel verbal and imaginal systems by which a given stimulus may be represented (2), the neural substrates of which have also been shown to be differently lateralized (3). Whereas previous encoding studies have examined variation across material types, the present study investigated hemispheric effects within material types. FMRI was used to determine the laterality of encoding-related activation for verbal stimuli that varied in concreteness, and for pictorial stimuli that varied in verbalizability. Methods Task Design Verbal task. Nine healthy, right-handed native English speakers viewed blocks of serially presented English nouns, alternating with blocks of control strings (2500ms presentation, 500ms ITI). Noun stimuli comprised two sublists: 40 concrete and 40 abstract nouns (mean concreteness ratings 6.31 and 2.61, respectively, on a 1-to-7 point scale (Toglia & Battig, 1978)). Each task block comprised ten nouns: eight from one sublist and two from the other. Subjects classified nouns as concrete or abstract, responding via a left-or-right button press. Control blocks comprised ten strings of Ls or Rs, eliciting the same proportion of left and right button presses as the preceding block of nouns. Subjects were instructed to remember noun stimuli for a post-scan forced-choice recognition memory test. Pictorial task. For each of two face memory encoding tasks, eleven healthy, right-handed volunteers viewed blocks of unfamiliar face photographs, alternating with blocks of a repeatedly presented pixelated control image (six 40s task/control blocks, 10 stimuli per block, 3500ms presentation, 500ms ITI). For the first task, full-head photographs were shown, including hair, neck, and upper shoulders. In some cases, clothing and jewelry were visible. For the second task, the same set of face photographs was used, but each photograph was cropped so as to include the brow, eyes, nose, and mouth, but exclude ears, hair, and any extraneous objects. Subjects were instructed to remember the faces for a post-scan recognition test, and to attend the control images but not to memorize them. Scanning occurred during the encoding tasks but not during recognition testing. Image Acquisition and Data Analysis BOLD functional images were obtained at 1.5Tesla in 20 contiguous 5-mm-thick axial slices. Multisubject SPMs were constructed in Talairach space using the SPM99{t} random effects model. Cognitive subtraction revealed activation associated with encoding during blocks of predominantly concrete and predominantly abstract nouns, and during cropped-face and full-head encoding. For each task, activation exceeding a statistical threshold (∝=.05) was quantified in two a priori-defined regions of interest. ROIs comprised the inferior frontal gyrus (IFG) and fusiform gyri (FG), as these structures have demonstrated reliable material- specific effects in previous encoding studies. Hemispheric asymmetry of activation in each ROI was assessed using an asymmetry ratio [AR = (VoxelsR– VoxelsL)/(VoxelsR+VoxelsL)]. Results and Discussion Differing hemispheric effects shown previously across verbal and pictorial material types were demonstrated in the present study within material types. Average activation in the ROIs was bilateral during encoding of concrete nouns [AR(IFG)=-.06, ns; AR(FG)=.13, ns], which are amenable to both verbal an imaginal coding, but significantly left-lateralized during encoding of abstract nouns [AR(IFG)=-.22, P=.001; AR(FG)=-.80, P=.009], which are resistant to imaginal coding. Activation during encoding of full-head photographs was left-lateralized in the IFG [AR(IFG)=-.46, P=.001] and bilateral in the FG [AR(FG)=-.25, ns], but activation during encoding of cropped-faces, which are resistant to verbal coding, was significantly right- lateralized in both ROIs [AR(IFG)=.12, P=.001; AR(FG)=.71, P=.001]. Replication of these findings within verbal stimuli that vary in imageability and within pictorial stimuli that vary in verbalizability would suggest that hemispheric specialization during memory encoding heretofore described as material- specific might be more accurately described as code- specific. References Casasanto, D.J., et al. (2000) in Proceedings of the Cognitive Science Society 22, 77-82. Paivio, A. (1991) Canadian Journal of Psychology 45, Kounios, J. & Holcomb, P.J. (1994) J. Exp. Psych.: Learning, Memory, & Cognition 20, 804-823.
The author elaborates on the meanings of the adjectives „srpski“ and \n„srbijanski“ in general and in the phrases where both these adjectives refer to the \nrepublic of Serbia (both country and state) in particular. The latter can be seen in \nthe examples such as srpska (or srbijanska) vlada the government of Serbia. The \nauthor takes into consideration the existing body of literature about this subject \n(authors such as: Е. Fekete, М. Nikolič, І. Кlajn, and М. Sipka) and analyzes the \ninstances of ambiguity involving „srpski“ and „srbijanski“ (as in the example abоve). Using the lexical and semantic norms of the contemporary standard Serbian \nlanguage, the author points to the possibility of differentiating the usage of these \ntwo adjectives. Нe also emphasizes the need of further elaboration and verification \nof the linguistic standardization criteria in general and in the field of the lexicon \n(lexical meaning and usage) in particular.
Bien loin que l'objet precede le point de vue, on dirait que c'est le point de vue qui cree l'objet (Ferdinand de Saussure 1916:23).Traditionally, historiographers of Afrikaans have argued that a relatively uniform (spoken) vernacular existed at the Cape from the late eighteenth century, where it constituted the L(ow) variety in a diglossic situation (see Raidt 1991). Standardization of Afrikaans has been described accordingly in a “naturalistic” fashion as the codification and elaboration of this preexistent vernacular. This paper summarizes the results of a variationist study of late nineteenth- and early twentieth-century documents that shows the traditional view to be seriously flawed. While “Cape Dutch Vernacular” (or “Afrikaans,” as it came to be called) was a well-defined entity in the popular consciousness from the mid-nineteenth century, the actual patterns of language use in the historical texts indicate the existence of a complex social dialect continuum until the early twentieth century. Variation patterns described for the late eighteenth century are shown still to be productive around 1900. Linguistic standardization, understood as the reduction of variation and the emergence of a linguistic norm, was rapid and strongly marked by cultural and political nationalism.* I should like to thank Roger Lass and Paul Roberge for many stimulating discussions about Afrikaans historical linguistics and for their comments on earlier versions of the work presented in this article. The responsibility for any mistakes remains of course my own.
espanolEn este articulo se examina el uso de ir a + infinitivo y del futuro en -re en las doce ciudades que se incluyen en el Macrocorpus de la norma linguistica culta de las principales ciudades del mundo hispanico. El analisis cuantitativo proporcionara la frecuencia de empleo de la perifrasis y de la forma en -re cuando se utilizan para expresar futuridad, asi como las diferencias que puedan existir entre las distintas ciudades. Tambien se estudiara la incidencia del tipo de oracion, el sexo y la edad en el uso de ambas formas. Finalmente, el analisis de regresion multiple nos permitira establecer la probabilidad de que los condicionantes analizados condicionen de manera significativa la eleccion de las dos formas de futuro. EnglishThis paper examines the use of the verb group 'ir a + infinitive' and the future-tense form ending in '-re' in the twelve cities included in the Macrocorpus de la norma linguistica culta de las principales ciudades del mundo hispanico (The Macrocorpus of the cultural linguistic norm of the majar cities of the Hispanic World). The quantitative analysis will provide results conceming how often the verb group 'ir a + infinitive' and the '-re' form are employed to express future time, as well as the differences between the cities reviewed. Sentence-type, sex and age will be other variables whose frequency will also be studied with respect to the use of both forms. Finally, the multiple regression analysis will allow us establish how likely the aforementioned variables will significantly condition the choice of these future-tense forms.
Manual, large scale (computational) grammar development is time consuming, expensive and requires lots of linguistic expertise. More recently, a number of alternatives based on treebank resources (such as Penn-II, Susanne, AP treebank) have been explored. The idea is to automatically ``induce'' or rather read off (P)CFG grammars from the parse annotated treebank resources and to use the treebank grammars thus obtained in (probabilistic) parsing or as a starting point for further grammar development. The approach is cheap, fast, automatic, large scale, ``data driven'' and based on real language resources.\n\nTreebank grammars typically involve large sets of lexical tags and non-lexical categories as syntactic information tends to be encoded in monadic category symbols. They feature flat rules (trees) that can ``underspecify'' attachment possibilities. Treebank grammars do not in general follow Xbar architectural design principles (this is not to say that treebank grammars do not have design principles). As a consequence, treebank grammars tend to have very large CFG rule bases (e.g. Penn-II > 17,000 CFG rules for about 1 million words of text) with often only minimally differing rules. Even though treebank grammars are large, they are still incomplete, exhibiting unabated rule accession rates. From a grammar engineering point of view, the size of the rule base poses problems for maintainability, extendability and, if a treebank grammar is to be used as a CF-base in a LFG grammar, for functional (feature-structure) annotations. From the point of view of theoretical linguistics, flat treebank trees and treebank grammars extracted from such trees do not express linguistic generalisations. From the perspective of empirical and corpus linguistics, flat trees are well-motivated as they allow underspecification of subtle and often time consuming attachment decisions. Indeed, it is sometimes doubted whether highly general Xbar schemata usefully scale to ``real'' language.\n\nIn previous work we developed methodologies for automatic feature-structure annotation of grammars extracted from treebanks. Automatic annotation of ``raw'' treebank grammars is difficult as annotation rules often need to identify subsequences in the RHSs of flat treebank rules as they explicitly encode head, complement and modifier relations. Xbar based CFG rules should substantially facilitate automatic feature-structure annotation of grammar rules.\n\nIn the present paper we conduct a number of experiments to explore a space of possible grammars based on a small fragment of the AP treebank resource. Starting with the original treebank fragment we automatically extract a CFG G. We then apply an automatic structure preserving grammar compaction step which generalises categories in the original treebank fragment and reduces the number of rules extracted, resulting in a generalised treebank fragment and in a compacted grammar Gc. The generalised fragment is then manually corrected to catch missed constituents (and the like) resulting in an automatically extracted, compacted and (effectively manually) corrected grammar Gc,m. Manual correction proceeds in the ``spirit'' of treebank grammars (we do not introduce Xbar analyses). We then explore how many of the manual correction steps on treebank trees can be achieved automatically. We develop, implement and test an automatic treebank ``grooming'' methodology which is applied to the generalised treebank fragment to yield a compacted and automatically corrected grammar Gc,a. Grammars Gc,m and Gc,a are very similar to compiled out ``flat'' LFG-82 style grammars. We explore regular expression based compaction (both manual and automatic) to relate Gc,m to a LFG-82 style grammar design. Finally, we manually recode a subsection of the generalised and manually corrected treebank fragment into ``vanilla-flavour'' XBar based trees. From these we extract a compacted, manually corrected, XBar based grammar Gc,m,x. We evaluate our grammars and methods using standard labelled bracketing measures and according to how well they perform under automatic feature-structure annotation tasks.
The morphology of inflectional languages poses specific problems in the processing of morphological alternations. Regular alternations at morpheme boundaries can be elegantly captured by the use of rule formalisms based on the two-level morphology model. Stem alternations and completely irregular alternations at morpheme boundaries, however, need to be captured in some way in the lexicon. This paper presents four possible solutions to the problem and makes a claim in favor of one of them. The proposed approach makes use of feature bundles that contain the necessary linguistic information to uniquely identify allomorphic variations of stems in the lexicon. The proposal is an improvement in that it simplifies the representation of allomorphic variations in the lexicon by avoiding duplication of stem allomorphs to capture cross-combination of several morphosyntactic features in stem+flex sequences.
This paper describes how traditional andnon-traditional methods were used to identifyseventeen previously unknown articles that webelieve to be by Stephen Crane, published inthe New-York Tribune between 1889 and1892. The articles, printed without byline inwhat was at the time New York City's mostprestigious newspaper, report on activities ina string of summer resort towns on New Jersey'snorthern shore. Scholars had previouslyidentified fourteen shore reports as Crane's;these possible attributions more than doublethat corpus. The seventeen articles confirmhow remarkably early Stephen Crane set hisdistinctive writing style and artistic agenda. In addition, the sheer quantity of the articlesfrom the summer of 1892 reveals how vigorouslythe twenty-year-old Crane sought to establishhimself in the role of professional writer. Finally, our discovery of an article about theNew Jersey National Guard's summer encampmentreveals another way in which Crane immersedhimself in nineteenth-century military cultureand help to explain how a young man who hadnever seen a battle could write so convincinglyof war in his soon-to-come masterpiece,The Red Badge of Courage. We argue that thejoint interdisciplinary approach employed inthis paper should be the way in whichattributional research is conducted.
Imagine discourse between the arts in which the conventions of what we might call ordinary cognition do not apply, on site of intense lobbying neither tethered by history or cultural integrity, nor, frequently, concerned with social cohesion or communicative norms. It will be discourse in which the categories of an imperial culture are abrogated (however temporarily) by an indigenous one, yet it will undoubtedly also be site of intense colonization. On it, likewise, there will be an appropriation of language on an unprecedented scale. Past experience will play little part. Memory short and episodic, rather than semantic. It primal discourse. Primal in that it the site of first contact. Primal also in that it most often considered be the meeting of primitive culture and an advanced. Primal, likewise, in behavioral sense: in it, the satisfaction of physiological needs tantamount. Indeed, body and mind here are in state of kinetic unrest. This scene of prolonged immat urity, yet ontological and epistemological questions held in private language are encouraged be made public. Here the verbal arts have no canon. Literature has no prevailing cultural standard of merit. Questions of the popular and the high cultural are not naturalized and the fictional and nonfictional carry the same degree of verisimilitude as works of propaganda, rhetoric, and didacticism. In modem times, the West has become the site of this tenacious yet frequently unacknowledged imperial discourse, the discourse between multifarious forms of artistic representation win the attention of children. It in such discourse that the picture book located. 1 Arguing the need for critical language for the discussion of children's picture books, Peter Hunt suggests that to pictures into the same mould as words seems be potentially unproductive, except in terms of establishing conventions, when, of course, it is, by definition, necessary (181). It impossible, however, conventionalize pictorial representation the same degree as linguistic representation. Linguistic systems are mastered painstakingly, piece by piece, referent by referent, word by word. Pictorial systems, by contrast, are mastered all at once; they involve what Flint Schier has called natural generativity and are therefore much less conventional than linguistic systems. Each system, nevertheless, relies on general agreement and on willingness engage in communicative activity: the pictorial system on deep recognitional capacities that link object and its picture, the linguistic system on lexical and syntactical regularities and rules. In the media-saturated culture of the contemporary West, the commonalities and differences of our separate but shared experiences are frequently offered up in televisual or hypertextual format in which what Hunt describes as force set of discursive practices that address and interpellate both adults and children as potential viewers or listeners. The linguistic and the pictorial are frequently experienced as synergistic or polylogic systems bound up in this mass media, media whose intention, according Jean Baudrillard, transcribe the complexity of contemporary life into an ongoing procession of meaningless simulacra, hyperreal, a real without origin or reality (2). Baudrillard's disenchanted vision of postmodernity, articulated most profoundly in the late 1970s and 1980s, produced an interesting ontological metaphor. Disneyland, he claimed, is there conceal the fact that it the 'real' country, all of 'real' America, which Disneyland (just as prisons are there conceal the fact that it the social, in its entirety, in its banal omnipresence, which carceral). Disneyland presented as imaginary in order make us believe that the rest real, when in fact all of Los Angeles and the America surrounding it are no longer real, but of the order of the hyperreal and of simulation (25). …
AIM: In this paper the balance of affective and instrumental communication employed by nurses during the admission interview with recently diagnosed cancer patients was investigated. RATIONALE: The balance of affective and instrumental communication employed by nurses appears to be important, especially during the admission interview with cancer patients. METHODS: For this purpose, admission interviews between 53 ward nurses and simulated cancer patients were videotaped and analysed using the Roter Interaction Analysis system, in which a distinction is made between instrumental and affective communication. RESULTS: The results reveal that more than 60% of nurses' utterances were of an instrumental nature. Affective communication occurred, but was more related to global affect ratings like giving agreements and paraphrases than to discussing and exploring actively patients feelings by showing empathy, showing concern and optimism. CONCLUSION: In future, nurses should be systematically provided with (continuing) training programmes, in which they learn how to communicate effectively in relation to patients' emotions and feelings, and how to integrate emotional care with practical and medical tasks.
This paper presents a bottom-up generator that makes use of Information Retrieval techniques to rank potential generation candidates by comparing them to a data base of stored instances. We introduce two general techniques to address the search problem, expectation-driven search and dynamic grammar rule selection, and present the architecture of an implemented generation system called IGEN. Our approach uses a domain-specific generation grammar that is automatically derived from a semantically tagged treebank. We then evaluate the efficiency of our system.
This document describes the Part-of-Speech (POS) tagging guidelines for the Penn Korean Treebank Project. The corpus used for this project consists of around 54,000 words and 5,000 sentences. This document starts with a summary of the tagset used in the Penn Korean Treebank, followed by a more detailed discussion of each tag with examples. Then pairs of tags that are easily confused with each other are discussed and guidelines on how to distinguish one from the other for a given base forms and inflections are presented. The document concludes with a list of specific problematic examples with guidelines on how to handle such cases.
Reviewed by: New horizons in the study of language and mind by Noam Chomsky D. Terence Langendoen New horizons in the study of language and mind. By Noam Chomsky. Cambridge: Cambridge University Press, 2000. Pp. xvii, 230. This is a collection of seven essays based on lectures and articles by Noam Chomsky from 1992 to the present, together with a foreword by Neil Smith. C has published a number of books like this one over the years, which attack the empiricist philosophy of language of Quine, Putnam, Davidson, and others and which defend his own ‘naturalist’ and ‘internalist’ views. This book also traces developments in the philosophy of language from the time of Sir Isaac Newton, and thus picks up where Cartesian linguistics (1966) leaves off. C points out that the problem of reconciling the ‘mental’ with the ‘physical’ was fundamentally altered by Newton’s demonstration that Cartesian mechanism is untenable. The ultimate solution to the ‘mind-body problem’, if it is found at all, is not likely to involve a reduction of the mental to the physical. Rather, the mental should be studied just like the physical, using whatever tools, methods, and insights are available, without arbitrary stipulations such as those of the philosophers mentioned above who limit the study of language in particular to correlations with observable behavior. Many, if not most linguists, C observes, ignore the strictures of these eminent philosophers, so that their efforts amount to nothing more than the harassment of the practitioners of an emerging science. [End Page 583] Since ‘natural language’ is what develops naturally in the course of language acquisition without instruction, the internalist and naturalist study of language does not consider those aspects of language which result from the imposition of community norms nor does it consider specialized uses which must be explicitly taught. For example, the common mass noun water does not mean ‘H2O’ in any natural language (thus rendering irrelevant to the study of natural languages such thought experiments as Putnam’s 1975 ‘twin earth’ thought experiments), and the consideration of what water does mean in a natural language leads to the conclusion that its reference cannot be determined extensionally. The same is true for every referring expression in a natural language, including proper nouns. Further, C maintains that the meanings of most lexical items in a natural language are far more elaborate than what is normally recorded in dictionaries and suggests that lexical structure is best explored within a decompositional framework such as that of Moravcsik 1990 or Pustejovsky 1995 as a kind of abstract syntax. In the first essay, C traces the evolution of his own conception of grammar, beginning with transformational-generative grammar in its various forms; continuing with the ‘principles and parameters’ framework, which he considers a more significant ‘revolution’ than transformationalgenerative grammar, the latter being a continuation of both traditional and structuralist ideas; and culminating in the ‘minimalist program’. In the principles and parameters framework, an internalized grammar (an I-language) is considered, in the words of the fifth essay, to be ‘an instantiation of the initial state [with the parameters fixed], idealizing from the actual states of the language faculty’, which are ‘the result of the interaction of a great many factors, only some of which are relevant to the inquiry into the nature of language’ (123). The development of the minimalist program was motivated by two closely related questions. First, ‘to what extent [can] the principles themselves... be reduced to deeper and natural properties of computation’ (123); and second, ‘to what extent [is] language... a “good solution” to the legibility conditions imposed by the external systems with which it interacts’ (9)? As the descriptor ‘minimalist’ suggests, C seeks a theory which is stripped to bare essentials. A language must contain phonetic and semantic features, a way of bundling these together into lexical items, and a way of combining lexical items together into larger expressions. It must also interact with other systems of the mind/brain which are responsible for producing and recognizing its expressions both phonetically and conceptually. An ideal or ‘perfect’ I-language is one whose computational apparatus consists only of entities and operations that are necessary to insure...
Results of the noun-verb pair comprehension and production tests from the Test Battery for Auslan Morphology and Syntax (Schembri et al., 2000) are re-presented, re-analyzed, and compared to data from two other cases also dealing with noun-verb pairs: the Auslan lexical database and a comparison of Auslan and American Sign Language (ASL) signs. The data elicited through the test battery and presented in this article confirm the existence of formationally related noun-verb pairs in Auslan in which the verb displays a single movement and the noun displays a repeated movement. The data also suggest that the best exemplars of noun-verb pairs of this type in Auslan form a distinct set of iconic (mimetic) signs archetypically based on inherently reversible actions (such as opening and shutting). This strong iconic link perhaps explains why the derivational process appears to be of limited productivity, though it does appear to have "spread" to a number of signs that appear to have no such iconicity. There appears to be considerable variability in the use of the derivational markings, particularly in connected discourse, even for signs of the "open and shut" variety. Overall, the derivational process is apparently still closely linked to an iconic base, is incipient in the grammar of Auslan, and is thus best described as only partially grammaticalized.
This paper presents a statistical parser for a wide-coverage Combinatory Categorial Grammar (CCG) derived from the Penn Treebank. The Treebank is translated to a corpus of canonical CCG derivations. We de ne a generative statistical model over CCG derivations and train it on the transformed Treebank.
The situation with regard to Russian language resources is fragmented and disorganized. For this reason, it is important to promote for Russian the development of its basic resources in one package that could be used for development of speech products. The paper presents a design of the Russian lexical databases, corpora and supporting tools (system for construction and support of lexical databases, system for transcription, morphological analyzer and normalyzer) developed for wide usage in speech engineering.
This article describes the challenges posed by optical musicrecognition – a topic in computer science that aims to convert scannedpages of music into an on-line format. First, the problem is described;then a generalised framework for software is presented that emphasises keystages that must be solved: staff line identification, musical objectlocation, musical feature classification, and musical semantics. Next,significant research projects in the area are reviewed, showing how eachfits the generalised framework. The article concludes by discussingperhaps the most open question in the field: how to compare the accuracy and success of rival systems, highlighting certain steps thathelp ease the task.