Papers reviewed and determined not to be word norm studies. Use the flag icon to report errors or suggest re-inclusion.
18625 papers
The suitability of computer‐based instruction (CBI) for workers with limited education was evaluated in an Hispanic orchard workforce that reported little computer experience and 5.6 mean years of formal education. Ladder safety training was completed by employees who rated the training highly (effect size [d_gain] = 5.68), and their knowledge of ladder safety improved (d_gain = 1.45). There was a significant increase (p < 0.01) in safe work practices immediately after training (d_gain = 0.70), at 40 days post training (d_gain = 0.87) and at 60 days (d_gain = 1.40), indicating durability. As in mainstream populations, reaction or affective ratings correlated well with utility ratings, but not with behavior change. This demonstrates that an agricultural workforce with limited formal education can learn job safety from CBI and translate the knowledge to work practice changes, and those changes are durable.
This paper presents an approach to dependency parsing which can utilize any standard machine learning (classification) algorithm. A decision list learner was used in this work. The training data provided in the form of a treebank is converted to a format in which each instance represents information about one word pair, and the classification indicates the existence, direction, and type of the link between the words of the pair. Several distinct models are built to identify the links between word pairs at different distances. These models are applied sequentially to give the dependency parse of a sentence, favoring shorter links. An analysis of the errors, attribute selection, and comparison of different languages is presented.
We present an automatic approach to tree annotation in which basic nonterminal symbols are alternately split and merged to maximize the likelihood of a training treebank. Starting with a simple X-bar grammar, we learn a new grammar whose nonterminals are subsymbols of the original nonterminals. In contrast with previous work, we are able to split various terminals to different degrees, as appropriate to the actual complexity in the data. Our grammars automatically learn the kinds of linguistic distinctions exhibited in previous work on manual tree annotation. On the other hand, our grammars are much more compact and substantially more accurate than previous work on automatic annotation. Despite its simplicity, our best grammar achieves an F1 of 90.2% on the Penn Treebank, higher than fully lexicalized systems.
In this paper we describe the current state of a new Japanese lexical resource: the Hinoki treebank. The treebank is built from dictionary definition sentences, and uses an HPSG based Japanese grammar to encode both syntactic and semantic information. It is combined with an ontology based on the definition sentences to give a detailed sense level description of the most familiar 28,000 words of Japanese.
Aim: The present study aims to understand environmental problems perception related to affectivity. Introduction and Context: The seriousness of global environmental problems engenders a challenge to psychology regarding the need to induce environmentally responsible behavior and sustainable development. Environmental psychology research has classically been cognitively oriented. Recent studies on risk perception, however, highlight the important role that affect may be playing, and sustainable behavior seems to stem from emotions (feelings of connectedness with other beings or moral emotions such as indignation or guilt).Method: 170 undergraduates where asked to recall associations for three environmental problems selected from a previous survey: climate change, loss of biodiversity, and environmental unconsciousness. Affective images associated with these issues were collected through free association test. Affect was also measured using the validated Spanish version of The Positive and Negative Affect Schedule (PANAS). One assumption of the current research is that word association techniques may allow to explore possible links between imagery and behavior. Despite the fact that this measuring instrument has recently been applied to the analysis of environmental risks, our methodological contribution is its combination with one of the most widely used measure of affectivity. That is to say, the second assumption rests on evidence suggesting a two dimensional structure of affective experience. 1509 associations were generated through the imagery survey and the content of these associations was analysed for each inductive stimulus.Affective imagery: The primary image of the majority of respondents was a paraphrase of each environmental issue. Although very few associations to climate change denoted consequences on human health, and no one mentioned future generations, our findings indicate a strong tendency for respondents to specify effects such as disasters, ice melting or floods, rather than causes, such as greenhouse gas emissions. By the same token, loss of biodiversity was primarily associated with resource scarcity or devastation, not with excessive uses of ecosystems. A slight difference appeared in images associated with environmental unconsciousness: the second highest frequency of responses referred to a specific behavioral pattern (no recycling actions), and the following ones suggested a sense of personal responsibility for environmental outcomes. In addition to analyzing the content categories, the affective ratings of images were examined.Implications: The knowledge gained from this research may be applied to the construction of communication messages and although more research is needed to compare data from different socio-demographic groups, there is some evidence that affective dimensions may be predictive of participation in pro-environmental behaviors.
Many natural language processing tasks make use of a lexicon – typically the words collected from some annotated training data along with their associated properties. We demonstrate here the utility of corpora-independent lexicons derived from machine readable dictionaries. Lexical information is encoded in the form of features in a Conditional Random Field tagger providing improved performance in cases where: i) limited training data is made available ii) the data is case-less and iii) the test data genre or domain is different than that of the training data. We show substantial error reductions, especially on unknown words, for the tasks of part-of-speech tagging and shallow parsing, achieving up to 20 % error reduction on Penn TreeBank part-of-speech tagging and up to a 15.7 % error reduction for shallow parsing using the CoNLL 2000 data. Our results here point towards a simple, but effective methodology for increasing the adaptability of text processing systems by training models with annotated data in one genre augmented with general lexical information or lexical information pertinent to the target genre (or domain). 1.
CT resulted in variable functional and structural changes in dementia, and conclusions are limited by heterogeneity and study quality. Larger, more robust studies are required to correlate these findings with clinical benefits from CT.
Automatic analysis of syntax is one of the core problems in natural language processing. Despite significant advances in syntactic parsing of written text, the application of these techniques to spontaneous spoken language has received more limited attention. The recent explosive growth of online, accessible corpora of spoken language interactions opens up new opportunities for the development of high accuracy parsing approaches to the analysis of spoken language. The availability of high accuracy parsers will in turn provide a platform for development of a wide range of new applications, as well as for advanced research on the nature of conversational interactions. One concrete field of investigation that is ripe for the application of such parsing tools is the study of child language acquisition. In this thesis, we describe an approach for analyzing the syntactic structure of spontaneous conversational language in parent-child interactions. Specific emphasis is placed on the challenge of accurately annotating the English corpora in the CHILDES database with grammatical relations (such as subject, objects and adjuncts) that are of particular interest and utility, to researchers in child language acquisition. This work involves rule-based and corpus-based natural language processing techniques, as well as methodology for combining results from different parsing approaches. We present novel strategies for integrating the results of different parsers into a system with improved accuracy. One practical application of this research is the automation of language competence measures used by clinicians and researchers of child language development. We present an implementation of an automatic version of one such measurement scheme. This provides not only a useful tool for the child language research community, but also a task-based evaluation framework for grammatical relation identification. Through experiments using data from the Penn Treebank, we show that several of the techniques and ideas presented in this thesis are applicable not just to analysis of parent-child dialogs, but to parsing in general.
With increasing numbers of Web users, there is a necessity to improve their Web site navigation experience over the Internet and a range of Web applications have emerged recently for this purpose. Many researchers have stressed the importance of identifying semantic relatedness of Web pages in such Web applications as Web site navigation, automatic tour generation and adaptive Web applications. One approach to identifying semantic relatedness between documents is to use lexical databases and lexical chains. For example, an approach using lexical chains has been proposed by Green for identifying paragraph similarity in a document [1]. However, due to the unacceptable length of time needed for lexical chaining and the difficulty of global representation of documents, Green used synset weight vectors to compare semantic relatedness between two documents. But his approach to identifying paragraph similarities can be extended to identify semantic similarities between documents. In this study, an approach to identifying semantic similarity between Web pages incorporating weighted lexical chains (SRWLC) and document properties based on reiteration, density, length and semantic distance is proposed. The two approaches (the proposed approach and Green’s approach - SR Green ) were empirically compared by determining the semantic relatedness of Web pages using human subjects. The research hypothesis of this research is that the proposed approach identifies significantly more semantically related pages with a higher precision than the approach that has been proposed by Green. The null hypothesis is that there is no significant different in identification of semantically related pages between the two approaches. In this context precision is defined as the proportion of retrieved pages that are relevant. Web pages belonging to the Department of Computer Science, Keele University are used for the empirical evaluation of the two methods. The semantic relatedness of all pages was identified using both approaches and a Web-based page categorisation exercise using human subjects was carried out for this empirical evaluation. The evaluation is Web based, and therefore can be carried out on the subjects’ preferred Web browser at his or her preferred time & place. Therefore the distortion effects are minimised and the results of the evaluation are realistic and also can be reliably generalised to some extent. An invitation e-mail was sent to twenty subjects during the first week of October 2004 giving a link and guidelines for them to start the experiment. Once the link on the email was clicked, subjects were shown the initial Web page, giving an introduction to the experiment and instructions on how to continue. Twelve out of twenty invited subjects completed the experiment. Two subjects attempted the experiments but couldn’t finish because of network problems; another three subjects couldn’t finish because of time restrictions, and the other three subjects did not respond at all. Therefore responses from only twelve subjects are used for experimental evaluation. The Wilcoxon signed ranks test returns a p value of 0.004, indicating that the null hypothesis can be rejected, and that there is evidence to suggest that the SR WLC approach identifies significantly more semantically-related pages with higher precision than the SR Green approach. The SRWLC approach should be evaluated further using different Web site contents and language styles (e.g. American/British English). It would be interesting to use more subjects from different backgrounds to do the evaluation. This would determine whether the results of the evaluation are influenced by the human subject’s background, such as their status (student, staff, or other) his familiarity with the pages of the test database, gender differences or the level of English knowledge. Most importantly the approach is believed to be valid for more general browsing environments than a computer science Website and a wider study is desirable.
Literature suggests that relatively simple stimuli such as emotional facial expressions elicit neural activation in subcortical-limbic regions whereas contextually richer emotional pictures generate activation in a broader network of prefrontal as well as subcortical-limbic regions. The extent to which contextual features modulate subjective and neural responses associated with responses to emotional faces is unclear. Normative valence and arousal ratings for a large corpus of affective pictures (IAPS) were reviewed to explore whether emotional pictures containing both faces and context evoked more intense subjective emotional reactions than faces presented alone. This review study demonstrated that subjective emotional reactions to emotional faces with contextual information were greater than those to faces. An fMRI study was conducted to examine neural reactivity to these two types of emotional stimuli. Eleven healthy right-handed subjects viewed passively emotional stimuli during event-related functional magnetic resonance imaging (fMRI) assessment. Emotional faces augmented by contextual information elicited significant brain activity in the prefrontal cortex (BA10/11/47) as well as amygdala and thalamus. In contrast, emotional facial expressions provoked neural responses only in the subcortical-limbic/paralimbic regions including amygdala, thalamus, insula and posterior cingulate gyrus. These findings suggest that there are different but overlapping brain networks engaged by emotional faces and faces augmented by contextual information. The amygdala and thalamus can be regarded as common regions associated with emotional processing. Prefrontal regions may be unique in more cognitive and conscious processing of emotional faces augmented by contextual information.
Chinese word segmentation and Part-of-Speech (POS) tagging have been commonly considered as two separated tasks. In this paper, we present a system that performs Chinese word segmentation and POS tagging simultaneously. We train a segmenter and a tagger model separately based on linear-chain Conditional Random Fields (CRF), using lexical, morphological and semantic features. We propose an approximated joint decoding method by reranking the N-best segmenter output, based POS tagging information. Experimental results on SIGHAN Bakeoff dataset and Penn Chinese Treebank show that our reranking method significantly improve both segmentation and POS tagging accuracies. 1
This paper presents an approach to the question whether it is possible to construct a parser based on ideas from case-based reasoning. Such a parser would employ a partial analysis of the input sentence to select a (nearly) complete syntax tree and then adapt this tree to the input sentence. The experiments performed on German data from the Tüba-D/Z treebank and the KaRoPars partial parser show that a wide range of levels of generality can be reached, depending on which types of information are used to determine the similarity between input sentence and training sentences. The results are such that it is possible to construct a case-based parser. The optimal setting out of those presented here need to be determined empirically.
Objective:This study aimed to examine the effects of haloperidol and amphetamine on human startle response modulated by emotionally-toned film clips. Method: Sixty participants, in two groups (one receiving haloperidol and the other receiving amphetamine) were tested using electromyography (EMG) to measure eye-blink muscle (orbicular oculi) while different emotions were induced by six 2-minute film clips. Results: An affective rating shows the negative and positive effects of the two drugs on emotional reactivity, neither amphetamine nor haloperidol had any impact on the modulation of the startle response. Conclusion: The methodological and theoretical aspects of the study and findings will be discussed.
The years 1940–2000 have been a period of uneven growth and development for Irish-language prose and drama. The most significant development of all during these decades was the emergence of a great diversity of literary forms as many writers gradually distanced themselves from the more restricting aspects of Revivalist ideology and began to experiment more with language, structure and subject matter. Critical perspectives also evolved and it was recognised that the cultural context in which the Irish language had survived now demanded expression in a manner which often defied the strictures imposed by genre or particular aesthetic paradigms. Linguistic norms are challenged in many works, both by Gaeltacht and non-Gaeltacht writers, and questions of cultural change and cultural hybridity become stylistic concerns as much as themes in much of the prose literature of the period. During this time Máirtín Ó Cadhain was to gain prominence as a major prose writer, and a host of other new and distinct voices emerged such as those of novelists and short story writers Diarmaid Ó Súilleabháin, Alan Titley, Micheál Ó Conghaile and Pádraig Ó Siadhail. These years also witnessed a continuation of a strong Gaeltacht tradition of autobiography and auto-ethnography, side by side with an unprecedented amount of experimental prose writing. The publishing house Sáirséal agus Dill, which was established in 1946, was responsible for the publication of much of the most original and innovative fiction of the period, while in the 1980s the newly established Coiscéim and Cló Iar-Chonnachta were to play a prominent role in the encouragement of many new and younger writers.
Consistency of corpus annotation is an essential property for the many uses of annotated corpora in computational and theoretical linguistics. While some research addresses the detection of inconsistencies in part-of-speech and other positional annotation (van Halteren, 2000; Eskin, 2000; Dickinson and Meurers, 2003a), more recently work has also started to address errors in syntactic and other structural annotation (Dickinson and Meurers, 2003b, 2005; Ule and Simov, 2004; Dickinson, 2005). Spoken language differs in many respects from written language, but to the best of our knowledge the issue of detecting errors in the annotation of spoken language corpora has not yet been systematically addressed. This is significant since spoken data is increasingly relevant for linguistic and computational research—and such corpora are starting to become more readily available, as illustrated by the holdings of the Linguistic Data Consortium (http://www.ldc.upenn.edu). This paper addresses the issue, based on the variation n-gram error detection approach developed in Dickinson and Meurers (2003a). We use the German Verbmobil treebank (Hinrichs et al., 2000) as an exemplar of a spoken language corpus and discuss properties of such corpora which are relevant when adapting the variation n-gram approach for detecting errors in syntactic annotation of spoken language corpora.
Previous articleNext article FreeCurrent ApplicationsLinguistic AnthropologyV.ChandV.Chand Search for more articles by this author PDFPDF PLUSFull Text Add to favoritesDownload CitationTrack CitationsPermissionsReprints Share onFacebookTwitterLinked InRedditEmailQR Code SectionsMoreImmigration Practices in Belgium: African AsylumSeeker DiscourseJan Blommaert, a Belgian anthropologist currently at the Institute of Education, University of London, is known in Belgium as a public campaigner on immigration issues. He became involved in African asylumseekers rights in 1998, when the death of Semira Adamu during her forced repatriation provoked a public outcry over the implementation of Belgian immigration policies, in which more than 95% of applicants for asylum are rejected. Adamu had fled Nigeria in her teens to avoid entering into a polygamous marriage with a 65yearold man. Shackled at the ankles and vigorously resisting, she was suffocated by Belgian police attempting to restrain her.Poster for a 2003 commemoration of Adamu's death.View Large ImageDownload PowerPointAdamus death was a catalyst for the formation of new action groups, and existing organizations also became involved: churches opened their doors to asylum seekers, and NGOs such as OXFAM and the League of Human Rights campaigned for asylum seekers rights. Early on, Blommaert contributed to this organic collaborative effort: I gave tons of public lectures for any audience willing to listen, wrote opeds in major newspapers, campaigned with MPs close to the government, participated in public debates on these matters, and wrote expert articles for a wider audience. Additionally, he saw that his understanding of the legal and sociolinguistic issues was directly applicable to the problem. He was motivated, he explains, by awareness of the real stakes and real people involved.African asylum seekers face circumstances not shared with those from Europe and the Gulf because they come from wartorn areas with unclear state boundaries, may be illiterate and lack documentation, have long migration paths to Belgium, and do not share a language with immigration authorities. In particular, Blommaert points out, their choices of language, background texts, and genres for storytelling affect their chances of acquiring refugee status. African asylumseeker language is stereotypically filled with language mixing, language impurities, varying degrees of literacy, and different styles of storytelling and discourse. When asylum seekers present their stories, they are unaware of how the officials are judging them on their linguistic habits. The officials note these details and find them inconsistent with Belgian expectations; they judge the asylum seekers in terms of these nave choices and reject them. In many such cases, rejection is a matter of life or death for the refugees, given the risks associated with deportation and with repatriation into the home country that originally motivated them to seek asylum.Blommaert has mobilized an informal network of Belgian academics and institutions that has produced many academically informed public statements. Additionally, by documenting and analyzing oral immigration interviews with an eye to understanding the disconnect between asylum seekers presentations and immigration officials expectations, he has improved practice in the Immigration Department. He has used his analysis to train members of the department, helping to adjust views of what can be gathered in an immigration interview, to improve interview techniques, and to promote an awareness of variability in sociocultural linguistic norms and presentation styles.He continues to work as an official expert for Belgian legal, government, and security authorities, translating and analyzing documentation and corroboratory written texts provided for immigration procedures and advising on specific issues. He has been able to influence outcomes for some individual asylum seekers. His public campaigning has raised awareness of the issues surrounding African asylum seekers, and his publications have provoked work within academia that may lead to further interventions by academics with regard to the interview process.Blommaert hopes that public campaigning by NGOs and academic scholarship will promote continued dialogue with immigration officials, eventually producing policies and procedures that take into account the politics of migration and displacement and the way in which asylum seekers frame their life stories. Previous articleNext article DetailsFiguresReferencesCited by Current Anthropology Volume 47, Number 3June 2006 Sponsored by the Wenner-Gren Foundation for Anthropological Research Article DOIhttps://doi.org/10.1086/504161 Views: 287Total views on this site Citations: 1Citations are reported from Crossref PDF download Crossref reports the following articles citing this article:Kevin D. O'Gorman, Cailein Gillespie The mythological power of hospitality leaders?, International Journal of Contemporary Hospitality Management 22, no.55 (Jul 2010): 659–680.https://doi.org/10.1108/09596111011053792
The goal of this report is to summarize our experiments and present the final result of our participation in the CoNLL-X Shared Task 2006. The topic of this year's shared task was multi-lingual dependency parsing. The organizers have prepared 13 existing dependency treebanks so that they all comply to the same markup format. The training and test data for the languages differ in size, granularity and quality, but they have tried to even out differences in the markup format. No additional information is allowed to be used besides the provided training data, forcing the parser to be fully automatic and data-driven. Ideally, the same parser should be trainable for all languages, possibly by adjusting parameters. The main goal is to assign labeled dependency structure for all languages on held out test data, approximately 5 000 tokens for each language. The main metric for comparison of the different parsers of the participants is therefore labeled attachment score.
Functional Combinatory Categorial Grammar (FCCG) advances the field of combi- natory categorial grammars by enabling semantic dependencies to be determined directly from the syntactic derivation under the action of a small set of extraction rules. Pred- icates are extracted composably and can be used to apply semantic constraints during parsing. The approach is an alternative to that of classical CCG which requires (i) map- ping from categories to lambda expressions, (ii) a set of semantic transformation rules for unary combination, and (iii) an explicit �-reduction stage. GFCCG, a generalised form of the grammar, has previously been applied to situation assessment (McMichael, Jarrad, & Williams, 2006). In FCCG, combinators are largely distinguished by their semantic purpose. Unary combination is only used for preterminal-terminal transitions. Replacing unary type- raising and type-changing by their binary counterparts R and P tends to reduce parse ambiguity. Four other binary combinators are introduced to model various semantic phenomena: functional composition (F), modification (M), apposition (A) and copular modification (Q). Of the combinators of classical CCG, only binary coordination (&) is retained. The category is the natural feature structure for CCG, and we show how it may be extended to host semantic and parsing-related features compactly. The grammar is demon- strated by extracting an FCCG-annotated corpus from the Penn treebank using the com- binator calculator described in (Foreman & McMichael, 2004). Only fifty categories and 140 productions cover 99.7% of the extracted corpus, a substantially more efficient repre- sentation than previous conversions. We adopt a factored conditional statistical model, and provide an efficient iterative learning algorithm involving direct feedback of parsing errors that does not require enu- meration of the parse forest. While the best parser performances on data derived from the Penn treebank have come from fully lexicalized parsers, we have been motivated by the need to provide good coverage outside that relatively narrow training domain. We have therefore used a semilexicalised feature model, which does not, for example, contain bilexicalised features. Spurious ambiguity is controlled via probabilistic scoring coupled with agenda-based A* parsing. The parser uses an auxiliary queue prioritized by multi-tag probabilities for terminal node introduction. The parser's syntactic dependency F-score of 79.2% compares well with the best labelled syntactic dependency F-score for a fully lexicalised CCG parser tested on data consistent with the training domain of 84.6% (Clark & Curran, 2004b). The parser obtained a semantic extraction F-score of 83.1%.
Asian languages are far from most western-style in their non-separate word sequence especially Chinese. The preliminary step of Asian-like language processing is to find the word boundaries between words. In this paper, we present a general purpose model for both Chinese word segmentation and named entity recognition. This model was built on the word sequence classification with probability model, i.e., conditional random fields (CRF). We used a simple feature set for CRF which achieves satisfactory classification result on the two tasks. Our model achieved 91.00 in F rate in UPUC-Treebank data, and 78.71 for NER task. 1
Abstract During spoken word-recognition, bilinguals have been shown to access their two languages simultaneously. The present study examined effects of language proficiency and lexical status on parallel language activation. Language proficiency was manipulated by testing German-native and English-native bilingual speakers of German and English. Lexical status was manipulated by presenting target words that either overlapped in form across translation equivalents (cognate words) or did not overlap in form across translation equivalents (English-specific words). Participants identified targets (such as hen) from picture-displays that also included similar-sounding German competitor words (such as Hemd, “shirt”). Eye-movements to German competitors were used to index co-activation of German. Results showed that both bilingual groups co-activated German while processing cognate targets; however, only German-native bilinguals co-activated German while processing English-specific targets. These findings indicate that high language proficiency and cognate status boost parallel language activation in bilinguals. Acknowledgements This work was supported by grants NICHD 1R03HD046952-01A1 and NSF BCS-0418495 to the second author, and by a Northwestern University Graduate Research Grant to the first author. Thanks go to Margarita Kaushanskaya, James Booth, Cynthia Thompson, and Judith Kroll for insightful comments and discussion, and to Valerie Burt, Gayatri Menon, Naveen Malik, Vridhi Chhabria, Nicole Kaligeropolous, Nadia Cone, Olga Boukrina, and Avital Rabin for their assistance at various stages of the project. We are grateful to Manuel Carreiras and to two anonymous reviewers for helpful comments and feedback on a previous version of this paper. Parts of this work were presented at the Fifth International Symposium on Bilingualism, and at the 27th Annual Meeting of the Cognitive Science Society. Notes 1For a linking hypothesis between linguistic processing and eye-movements, see Tanenhaus, Magnuson, Dahan, and Chambers (Citation2000). 2This translation was made by a fluent German–English bilingual, and back-translated to English by two other fluent German–English bilinguals for reliability. The German and English versions were then balanced by-item on word frequency [CELEX lexical database, Baayen, Piepenbrock, & Van Rijn, Citation1995, t (203) = 0.6, p>.5]. 3For purposes of subsequent analyses, English recordings of target words were compared to German recordings of competitor words (German recordings were made by native speakers of German). Onset similarity between English target-words and German competitor-words was assessed at the acoustic level (Ju & Luce, 2004) using Sound Studio software. The duration of acoustic overlap between target and competitor onsets was measured, and is shown in Table 2. A one-way ANOVA yielded significant differences between durations of target-competitor overlap in the low-, medium- and high-overlap conditions (cognate target condition, F(2, 30) = 45.5, p<.001; English target condition, F(2, 28) = 45.2, p<.001). Planned post-hoc comparisons yielded differences between low- and medium-overlap conditions, as well as between medium- and high-overlap conditions and between low- and high-overlap conditions (LSD post-hocs: p<.01). Significant differences between the three conditions were also found in terms of phonemes (cognate target condition, F (2, 30) = 18.6, p<.001; English target condition, F (2, 28) = 4.8, p<.05, and in terms of phonetic features: cognate target condition, F (2, 30) = 10.2, p<.001; English target condition, F (1, 28) = 6.1, p<.01. 4Of these, the target-competitor pairs cylinder / Zylinder (top-hat) and lock / Lockenwickler (curlers) were excluded due to within-language competition. The picture of curlers contained locks (of hair) and the top hat was cylinder-shaped. For evidence on eye-movements due to shape-similarities, see Dahan & Tanenhaus, 2005. The target-competitor pairs bear / Band (ribbon), file / Pfeil (arrow), turtle / Tuer (door) and tooth / Tuch (cloth) were excluded because these competitor pictures were fixated more than control items in the monolingual control group. 5By-item analyses were significant or marginally significant in all between-group comparisons, but were not significant in within-group comparisons. This is likely due to the fact that duration of overlap (in milliseconds) was a continuous variable. Variations on the continuum within each category were likely to elicit different responses, resulting in large standard errors. This variability among items contradicted the inherent assumption in by-item analyses that all items be similar, prompting analyses of variance by subject only. To verify effects of phonological overlap across items, regression analyses were performed (where degree of phonological overlap was regressed on likelihood of parallel activation). 6Follow-up t-tests were only conducted to compare percentages of looks to competitor vs. control items, with competitor-control item differences reflecting parallel language activation. Looks to competitors in one condition vs. another (or to control items in one condition vs. another) were not compared statistically, because they do not reflect parallel language activation. Any differences between looks to competitors in different conditions (or looks to control items in different conditions) may be due to the presence of different stimuli across these conditions. Further, differences between looks to competitors in different groups (or looks to control items in different groups) may be due to general differences in response latencies across groups. Follow-up t-tests were 1-tailed. 7Alternatively, it may be the case that the German L1 cohort was highly co-activated, even with target-competitor low phonological overlap. The increased duration of competition may have had no additional effect on competitor activation, and no sensitivity to phonological overlap would have been observed. Such an account has to be questioned for two reasons. First, absence of phonological overlap effects should also be apparent in monolingual auditory word recognition; however, this is not the case. Monolingual priming studies suggest that degree of phonological overlap influences co-activation of two words (Slowiaczek & Hamburger, 1992). Further, Weber and Cutler (2004) found a target-competitor overlap effect in Dutch-native bilinguals who processed words with confusable vowels in their second language, English. Therefore, since phonological overlap effects have been found within native and non-native languages, the absence of phonological overlap sensitivity, due to overall high L1 co-activation, is not a plausible account.
The Dutch spelling system, like other European spelling systems, represents a certain balance between preserving the spelling of morphemes (the morphological principle) and obeying letter-to-sound regularities (the phonological principle). We present experimental results with artificial learners that show a competition effect between the two principles: adhering more to one principle leads to more violations of the other. The artificial learners, memory-based learning algorithms, are trained (1) to convert written words to their phonemic counterparts and (2) to analyze written words on their morphological composition, based on data extracted from the CELEX lexical database. As an exception to the competition effect we show that introducing the schwa as a letter in the spelling system causes both morphology and phonology to be learnt better by the artificial learners. In general we argue that artificial learning studies are a tool in obtaining objective measurements on a spelling system that may be of help in spelling reform processes.
In this work we learn clusters of contextual annotations for non-terminals in the Penn Treebank. Perhaps the best way to think about this problem is to contrast our work with that of That research used treetransformations to create various grammars with different contextual annotations on the non-terminals. These grammars were then used in conjunction with a CKY parser. The authors explored the space of different annotation combinations by hand. Here we try to automate the process -to learn the "right" combination automatically. Our results are not quite as good as those carefully created by hand, but they are close (84.8 vs 85.7).
Standard techniques used in multilingual terminology management fail to describe legal terminologies as they are bound to different legal systems and terms do not share a common meaning. In the LexALP project, we use a technique defined for general lexical databases to achieve cross language interoperability between languages of the Alpine Convention. In this paper we present the methodology and tools developed for the collection, description and harmonisation of the legal terminology of spatial planning and sustainable development in the four languages of the countries of the Alpine Space.
This paper describes a parser which generates parse trees with empty elements in which traces and fillers are co-indexed. The parser is an unlexicalized PCFG parser which is guaranteed to return the most probable parse. The grammar is extracted from a version of the PENN treebank which was automatically annotated with features in the style of The annotation includes GPSG-style slash features which link traces and fillers, and other features which improve the general parsing accuracy. In an evaluation on the PENN treebank Its results for the empty category prediction task and the trace-filler coindexation task exceed all previously reported results with 84.1% and 77.4% fscore, respectively.
The present paper proposes a method by which to translate outputs of a robust HPSG parser into semantic representations of Typed Dynamic Logic (TDL), a dynamic plural semantics defined in typed lambda calculus. With its higher-order representations of contexts, TDL analyzes and describes the inherently inter-sentential nature of quantification and anaphora in a strictly lexicalized and compositional manner. The present study shows that the proposed translation method successfully combines robustness and descriptive adequacy of contemporary semantics. The present implementation achieves high coverage, approximately 90%, for the real text of the Penn Treebank corpus.
Treebank data have been utilized as data sources for a wide range of tasks in computational linguistics, including statistical parsing, anaphora resolution, induction of valence lexica, etc. More recently, researchers have experimented with extracting semantic information from syntactically annotated data. Here, treebank data
Traditionally, orthographic variants have been modelled as different ways of spelling the same word - described at the level of the lexeme. But when inflection is taken into account, this runs into a problem: different citation forms have different inflectional paradigm - and orthographic variation does not merely affect the citation form, but the entire paradigm. The MorDebe database therefore models orthographic variation as a relation between distinct, yet still token-identical lexemes. This paper discusses the advantage of that approach, and the full set of practical problems that arose during the structural treatment of orthographic variation in the MorDebe database.
This paper describes the Chinese NomBank Project, the goal of which is to annotate the predicate-argument structure of nominalized predicates in Chinese. The Chinese Nombank extends the general framework of the English and Chinese Proposition Banks to the annotation of nominalized predicates and adds a layer of semantic annotation to the Chinese Treebank. We first outline the scope of the work by discussing the markability of the nominalized predicates and their arguments. We then attempt to provide a categorization of the distribution of the arguments of nominalized predicates. We also discuss the relevance of the event/result distinction to the annotation of nominalized predicates and the phenomenon of incorporation. Finally we discuss some cross-linguistic differences between English and Chinese. 1.
We introduce Talbanken05, a Swedish treebank based on a syntactically annotated corpus from the 1970s, Talbanken76, converted to modern formats. The treebank is available in three different formats, besides the original one: two versions of phrase structure annotation and one dependency-based annotation, all of which are encoded in XML. In this paper, we describe the conversion process and exemplify the available formats. The treebank is freely available for research and educational purposes. 1.
The Surrey Cross-linguistic Database on Deponency encodes information on the presence of morphological mismatches in a controlled sample of genetically and geographically diverse languages (based on the 100-language sample from the World Atlas of Language Structures). The database was created for the project 'Extended Deponency: The right morphology in the wrong place', funded by the Economic and Social Research Council under grant number RES000230375.