Papers reviewed and determined not to be word norm studies. Use the flag icon to report errors or suggest re-inclusion.
18625 papers
Valence acquisition is the task consisting in the automatic extraction (learning) of subcategorisation – or argument structure – from corpora. In this talk I will concentrate on two issues. The first issue is: how much linguistic annotation do we need for valence acquisition? Approaches range from linguistically lean, e.g., Brent’s 1993 proposal to infer valence information from co-ocurrences of verbs with pronouns and functional words, to more recent proposals to read valence off from lavishly annotated treebanks. I will present the results of some experiments from Polish suggesting that shallow (or partial) parsing may be as useful in this task as more difficult and less efficient deep parsing. The second issue concerns the evaluation of the results of automatic valence acquisition. The common methodology is to compare the automatic results to a manually constructed valence dictionary, but – again on the basis of some experiments carried out for Polish – I will point out various weaknesses of this methodology and argue for the more costly and, hence, less common corpus-based evaluation.
Treebanks, or syntactically annotated corpora, are an invaluable resource for the development and evaluation of syntactic parsers, as well as for empirical research on natural language syntax. Treebanks for Swedish have a long and venerable history, represented by the pioneering work on Talbanken (Einarsson,
The dual process theory proposes that evaluative conditioning is a form of learning distinct from Pavlovian conditioning and that it displays different functional characteristics such as not being subject to modulation. However, when assessed online as opposed to post-experimentally, modulation of evaluative conditioning by context change has been found in a contingency reversal procedure. Reversal of evaluative learning was found to be faster when trained in a different context rather than in the original training context. The present study addressed the question whether context change or instructions would affect the rate of reversal of evaluative learning and whether reversal learning would accelerate across repetitions. A picture-picture paradigm was used to expose participants to CS-US pairs and contingency was reversed three times during the experiment. Participants were required to provide online causal judgements and valence ratings after each set of 10 training trials. Context change, but not instructions, displayed a trend in affecting reversal of evaluative learning with participants displaying faster learning on trials immediately subsequent to contingency reversal. Instructions affected the reversal of contingency judgements. There was no evidence of acceleration across repetitions for either measure or manipulation.
Traditional Active Learning (AL) techniques assume that the annotation of each datum costs the same. This is not the case when annotating sequences; some sequences will take longer than others. We show that the AL technique which performs best depends on how cost is measured. Applying an hourly cost model based on the results of an annotation user study, we approximate the amount of time necessary to annotate a given sentence. This model allows us to evaluate the effectiveness of AL sampling methods in terms of time spent in annotation. We acheive a 77% reduction in hours from a random baseline to achieve 96.5% tag accuracy on the Penn Treebank. More significantly, we make the case for measuring cost in assessing AL methods.
Data Oriented Parsing is a natural language processing model that analyses new input based on past experience. The underlying idea is to extract a set of fragment-probability pairs from a given treebank and use these concrete experiences to construct new utterance analyses. Initially, probabilities were based on the fragments' relative frequency of occurrence. This estimator, however, was soon shown to be biased towards large corpus trees [8] and inconsistent [10]. To alleviate the effects of bias on performance a set of heuristic constraints was put in force. Other estimators addressing these issues have since then been proposed. This paper seeks to show that the most commonly used DOP estimators are in fact susceptible to strong size-sensitive bias effects and to present a new estimation algorithm that greatly reduces these effects of bias on performance without complicating the estimation process.
Conversation is one of the poetical and cultural key concepts in the writings of Mme de Stael. Departing from the contemporary understanding of the term, she outlines the conversational act as an essentially transgressive praxis that surpasses a classical canon of social, aesthetic, and linguistic norms. Highlighting a broad variety of musical, sensual, and pre-semantic aspects of language, she insists on the informal dynamics of conversation. To a certain extent, Mme de Stael claims, all conversation is improvisation, and it is within improvisation that she locates (and welcomes) the trigger of unsuspected twists and turns, the spirited, the emphatic, and even drunken quality of verbal and non-verbal exchange. The paper discusses the development of this concept in Mme de Stael's major novel, Corinne ou l'Italie. It argues that this novel is one of the first to delineate conversation as transference, shifting the focus of attention from the realm of intentionality to non-thematic qualities of communication.
Underspecification-based algorithms for processing partially disambiguated discourse structure must cope with extremely high numbers of readings. Based on previous work on dominance graphs and weighted tree grammars, we provide the first possibility for computing an underspecified discourse description and a best discourse representation efficiently enough to process even the longest discourses in the RST Discourse Treebank.
Within the STEVIN1 project Large Scale Syntactic Annotation of written Dutch (LASSY), a manually corrected treebank of 1 million words is constructed. Lassy is part of a series of annotation projects for modern written and spoken Dutch. More specifically, it is an extension of the D-Coi and CGN projects, 2 and constitutes
We present a robust parser which is trained on a treebank of ungrammatical sentences. The treebank is created automatically by modifying Penn treebank sentences so that they contain one or more syntactic errors. We evaluate an existing Penn-treebank-trained parser on the ungrammatical treebank to see how it reacts to noise in the form of grammatical errors. We re-train this parser on the training section of the ungrammatical treebank, leading to an significantly improved performance on the ungrammatical test sets. We show how a classifier can be used to prevent performance degradation on the original grammatical data.
In this paper, we describe our work on building a parallel treebank for a less studied and typologically dissimilar language pair, namely Swedish and Turkish. The treebank is a balanced syntactically annotated corpus containing both fiction and technical documents. In total, it consists of approximately 160,000 tokens in Swedish and 145,000 in Turkish. The texts are linguistically annotated using different layers from part of speech tags and morphological features to dependency annotation. Each layer is automatically processed by using basic language resources for the involved languages. The sentences and words are aligned, and partly manually corrected. We create the treebank by reusing and adjusting existing tools for the automatic annotation, alignment, and their correction and visualization. The treebank was developed within the project Supporting research environment for minor languages aiming at to create representative language resources for language pairs dissimilar in language structure. Therefore, efforts are put on developing a general method for formatting and annotation procedure, as well as using tools that can be applied to other language pairs easily. 1.
In contrast to previous stress research, studies concerning phobic disorders have never systematically investigated individual response differences between phobic participants integrating numerous different response measures. The aim of this article is to clarify the existence of significant individual response differences in psychophysiological responses (e.g., heart rate, skin conductance responses (SCR), corrugator, cortisol), subjective ratings (e.g., valence, arousal), and avoidance behavior in 46 spider phobic and 44 non-phobic women when exposed to 20 phobic and 20 neutral pictures. Previous studies that did not attend to individual response differences showed that, during phobic stimulation, phobic individuals have increased psychophysiological responses (heart rate, SCR, and corrugator responses), more negative valence rating, and more subjective arousal than non-phobic individuals. These results were confirmed by our data. With regard to individual response uniqueness, 1/3-2/3 of spider-phobic women with low responsiveness in heart rate, cortisol, and avoidance behavior were indistinguishable from non-phobic women during phobic stimulation. With SCR, corrugator EMG, and subjective ratings, no individual response uniqueness was found. Based on the findings, exposure therapy might be improved by tailoring interventions to individuals with a therapeutic focus on those psychophysiological measures that show the highest individual responsivity.
espanolEste articulo presenta un metodo de identificacion y clasificacion de la valencia y las emociones presentes en un texto. Para ello, se introduce un nuevo concepto denominado disparador de emocion. Inicialmente, se construye de forma incremental una base de datos lexica de disparadores de emocion asociados a la cultura con la que se quiere trabajar, basandose en tres teorias diferentes: la Teoria de la Relevancia de Pragmatica, la Teoria de la Motivacion de Maslow de Psicologia y la Teoria de Necesidades de Neef de Economia. La base de datos creada parte de un conjunto inicial de terminos y es ampliada con la informacion de otros recursos lexicos, como WordNet, NomLex y dominios relevantes. El enlace entre idiomas se hace por medio de EuroWordNet y se completa y adapta a diversas culturas con bases de conocimiento especificas para cada lengua. Tambien, se demuestra como la base de datos construida puede ser utilizada para buscar en textos la valencia (polaridad) y el significado afectivo. Finalmente, se evalua el metodo utilizando los datos de prueba de la tarea no 14 de Semeval Texto afectivo y su traduccion al espanol. Los resultados y las mejoras se presentan junto con una discusion en la que se tratan los puntos fuertes y debiles del metodo y las directrices para el trabajo futuro. EnglishThis paper presents a method to automatically spot and classify the valence and emotions present in written text, based on a concept we introduced-of emotion triggers. The first step consists of incrementally building a culture dependent lexical database of emotion triggers, emerging from the theory of relevance from pragmatics, Maslow's theory of human needs from psychology and Neef's theory of human needs in economics. We start from a core of terms and expand them using lexical resources such as WordNet, completed by NomLex, sense number disambiguated using the Relevant Domains concept. The mapping among languages is accomplished using EuroWordNet and the completion and projection to different cultures is done through language-specific commonsense knowledge bases. Subsequently, we show the manner in which the constructed database can be used to mine texts for valence (polarity) and affective meaning. An evaluation is performed on the Semeval Task No. 14: Affective Text test data and their corresponding translation to Spanish. The results and improvements are presented together with an argument on the strong and weak points of the method and the directions for future work.
This research shows a new approach and development of a design methodology, based on the perspective of meanings. In this study the design process is explored as a development of the structure of meanings. The processes of search and evaluation of meanings form the foundations of developing this structure. In order to facilitate the use and operation of the meanings, the WordNet lexical database and an existing visualization of WordNet — Visuwords — is used for the process of meaning search. The basic tool used for evaluation process is the WordNet::Similarity software, measuring the relatedness of meanings in the database. In this way it is measuring the degree of interconnections between different meanings. This kind of search and evaluation techniques are later on incorporated into our methodology of the structure of meanings to support the design process. The measures of relatedness of meanings are developed as convergence criteria for application in the processes of evaluation. Further on, the methodology for the structure of meanings developed here is used to construct meanings in a verification of product design. The steps of the design methodology, including the search and evaluation processes involved in developing the structure of the meanings, are elucidated. The choices, made by the designer in terms of meanings are supported by consequent searches and evaluations of meanings to be implemented in the designed product. In conclusion, the paper presents directions for developing and further extensions of the proposed design methodology.
One problem facing the extraction of treebank grammars is that of ad hoc rules, rules used for constructions specific to one data set and unlikely to be used on new data (Dickinson, 2008). These rules can be erroneous, cover ungrammatical text, or reveal issues with the treebank’s annotation scheme.
A treebank is a text corpus in which each sentence has been annotated with its syntactic structure. Although the construction of a treebank is an expensive task, we believe that it is indispensable for the development of real applications in the field of Natural Language Processing (NLP) and also for the development of the
We present an approach to creating a treebank of sentences using multiple notations or linguistic theories simultaneously. We illustrate the method by annotating sentences from the Penn Treebank II in three different theories in parallel: the original PTB notation, a Functional Dependency Grammar notation, and a Government and Binding style notation. Sentences annotated with all of these theories are represented in XML as a directed acyclic graph where nodes and edges may carry extra information depending on the theory encoded. 1.
Combining Statistical and Rule-Based Approaches to Morphological Tagging of Czech Texts This article is an extract of the PhD thesis (Spoustová, 2007) and it extends the article (Spoustová et al., 2007). Several hybrid disambiguation methods are described which combine the strength of hand-written disambiguation rules and statistical taggers. Three different statistical taggers (HMM, Maximum-Entropy and Averaged Perceptron) and a large set of hand-written rules are used in a tagging experiment using Prague Dependency Treebank. The results of the hybrid system are better than any other method tried for Czech tagging so far.
Specific phobias are the most common anxiety disorder and are characterized by avoidance behavior. Avoidance behavior impacts daily function and is proposed to impair extinction learning. However, despite its prevalence, its objective assessment remains a challenge. To this end, we developed a fully automated experimental procedure using immersive virtual reality. The procedure contained a behavioral search, forced-choice, and an approach task with varying degrees of freedom and task relevance of the stimuli. In this study, we examined the sensitivity and feasibility of these tasks to assess avoidance behavior in patients with specific phobia. We adapted the tasks by replacing the originally conditioned stimuli with a spider and a neutral animal and investigated 31 female participants composed of 15 spider-phobic and 16 non-phobic participants. As the non-phobics were quite heterogeneous in terms of their Fear of Spiders Questionnaire (FSQ) scores, we subdivided them into six "fearfuls" that had elevated FSQ scores, and 10 "non-fearfuls" that had no fear of spiders. The phobics successfully managed to complete the procedure and showed consistent avoidance behavior across all behavioral tasks. Compared to the non-fearfuls, which did not show any avoidance behavior at all, the phobics looked at the spider much more often and clearly directed their body toward it in the search task. In the approach task, they hesitated most when they were close to the spider, and their difficulty to touch the spider was reflected in a strong increase in right hand acceleration changes. The fearfuls showed avoidance behavior depending on the tasks: strongest in the search task and weakest in the approach task. Additionally, we identified subjective valence ratings of the spider as the main influence on both objective avoidance behavior and subjective well-being after exposure, mediating the effect of the FSQ. In summary, the behavioral tasks are well suited to assess avoidance behavior in phobic participants and provide detailed insights into the process of avoidance.
Sometreebanks, such as German TIGER/NeGra, represent discontinuous elements directly, i.e. trees contain crossing edges, but the context-free grammars that are extracted from them, fail to make any use of this information. In this paper, we present amethod for extracting mildly context-sensitive grammars, i.e. simple range concatenation grammars (RCGs), from such treebanks. A measure for the degree of a treebank’s mild contextsensitivity is presented and compared to similar measures used in non-projective dependency parsing. Our work is also compared to discontinuous phrase structure grammar (DPSG).
Graph-based and transition-based approaches to dependency parsing adopt very different views of the problem, each view having its own strengths and limitations. We study both approaches under the framework of beamsearch. By developing a graph-based and a transition-based dependency parser, we show that a beam-search decoder is a competitive choice for both methods. More importantly, we propose a beam-search-based parser that combines both graph-based and transitionbased parsing into a single system for training and decoding, showing that it outperforms both the pure graph-based and the pure transition-based parsers. Testing on the English and Chinese Penn Treebank data, the combined system gave state-of-the-art accuracies
An algorithm based on information retrieval that applies the lexical database WordNet together with a linear discriminant function is proposed. It calculates the degree of similarity between words and their relative importance to support the development of distributed applications based on web services. The algorithm uses the semantic information contained in the Web Service Description Language specifications and ranks web services based on their similarity to the one the developer is searching for. It is applied to a set of 48 real web services in five categories, then compared them to four other algorithms based on information retrieval, showing an averaged improvement over all data between 0.6% and 1.9% in precision and 0.7% and 3.1% in recall for the top 15 ranked web services. The objective was to reduce the burden and time spent searching web services during the development of distributed applications, and it can be used as an alternative to current web service discovery systems such as brokers in the Universal Description, Discovery, and Integration (UDDI) platform.
The progress of Chinese dependency treebank construction has fallen behind other languages, such as English, in terms of scale and quality. Building a large scale treebank needs a lot of human and material resources. Meanwhile, it is very difficult to guarantee the quality of the treebank. In this paper, we explore a new method which combines rule-based method and statistical-based method to convert a constituent treebank named Penn Chinese Treebank to a dependency treebank which follows the annatation standard of HIT Chinese Dependency Treebank (HIT-IR-CDT). We increase the size of training data by adding converted treebank into HIT-IR-CDT and re-train the dependency parser. Experiments show that small addition of converted treebank can improve the performance of dependency parser, while large addition will bring it down. Through detailed analysis, we believe that convertion of constituent-to-dependency treebank, being a method of improving performance of dependency parser by utilizing different treebanks, still needs in-depth research.
Abstract Large linguistic databases, especially databases having a global coverage, such as the World Atlas of Language Structures, the Automated Similarity Judgment Program, and Ethnologue, are making it possible to systematically investigate many aspects of how languages change and compete for viability. Agent‐based computer simulations supplement such empirical data by analyzing the necessary and sufficient parameters for the current global distributions of languages or linguistic features. By combining empirical datasets with simulations and applying quantitative methods, it is now possible to address fundamental questions, such as ‘what are the relative rates of change in different parts of languages?’, ‘why are there a few large language families, many intermediate ones, and even more small ones?’, ‘do small languages change faster or slower than large ones?’, or ‘how does the borrowing of words relate to the borrowing of structural features?’
Abstract Affective ratings of multiple religious (sub)groups (Muslims, Christians, Jews and non-believers, as well as Sunni, Alevi and Sjiit Muslims), the endorsement of Islamic minority rights and religious group identification were examined among Sunni and Alevi Turkish-Dutch participants. The findings show that both groups differ in important ways. Some Alevi participants considered themselves Muslims but others interpreted Alevi identity in a secular way. The Sunnis were quite negative towards Jews and non-believers, they more strongly endorsed Islamic minority rights and they had very high Muslim group identification. Furthermore, the Sunnis were negative towards Alevis and the Alevis were negative towards the Sunnis. Muslim group identification was positively and strongly related to feelings towards Muslims and to the endorsement of Islamic group rights.
The cycle of lexicographic and linguistic work involved in compiling a computational phraseological database is divided into three phases and described in relation to the specific challenges multi-word expressions (MWEs) pose for a lexical database. Data collection is a process that is far from complete for the MWEs found in English, with the variability of some phrases making identification of all occurrences in large corpora a major challenge. Formalization of the form and variability ofMWEs is an interrelated process which can improve tools for data collection and other applications. Increased use of the phraseological lexical database in NLP applications can ultimately lead to further insights into the nature of MWEs and to improvements in the database. Due to the volume of lexicographic data on MWEs that still needs to be collected, analysed and formalized, and the cyclical nature of the work, the resulting lexical database should be reusable in as many applications as possible. WordManager-PhraseManager, the lexical resource described in the second part of the chapter, can capture the variability ofMWEs in a way that allows for maximum reusability of lexical data.
Semantic information extraction is a process of extracting concepts, entities, relations and entailment rules from a document. We propose an approach to extract concepts, entities and relations from the domain specific Tamil textual documents corpus. Concepts are extracted out by identifying semantic relationship between unique terms in documents by lexical database namely WordNet. Entity extraction process identifies proper names and its predefined named entity categories. Relation between two entities or between two concepts or between concept and entity in each sentence are extracted. This extracted semantic knowledge is exploited as a resource for extracting semantic information from the domain-related documents.
This paper is a project report of the lexicographic Internet portal OWID, an Online Vocabulary Information System of German which is being built at the Institute of German Language in Mannheim (IDS). Overall, the contents of the portal and its technical approaches will be presented. The lexical database is structured in a granular way which allows to extend possible search options for lexicographers. Against the background of current research on using electronic dictionaries, the project OWID is also working on first ideas of useradapted access and user-adapted views of the lexicographic data. Due to the fact that the portal OWID comprises dictionaries which are available online it is possible to change the design and functions of the website easily (in comparison to printed dictionaries). Ideas of implementing user-adapted views of the lexicographic data will be demonstrated by using an example taken from one of the dictionaries of the portal, namely elexiko.
In this paper the role of concept characteristics in lexical dialectometric research is examined in three consecutive logical steps. First, a regression analysis of data taken from a large lexical database of Limburgish dialects in Belgium and The Netherlands is conducted to illustrate that concept characteristics such as concept salience, concept vagueness and negative affect contribute to the lexical heterogeneity in the dialect data. Next, it is shown that the relationship between concept characteristics and lexical heterogeneity influences the results of conventional lexical dialectometric measurements. Finally, a dialectometric procedure is proposed which downplays this undesired influence, thus making it possible to obtain a clearer picture of the ‘truly’ regional variation. More specifically, a lexical dialectometric method is proposed in which concept characteristics form the basis of a weighting schema that determines to which extent concept specific dissimilarities can contribute to the aggregate dissimilarities between locations.
The studies on lexicographic definitions connected with the French tradition take charge eminently of typology and leave aside the question of metalanguage. So, in lexicography, the metalinguistic definition is often considered in the typological frame. This is because the above-mentioned studies are mostly based upon definitions either of nouns or verbs. In my presentation I shall attempt to demonstrate, from defining statements of the syncategorematic words drawn from the Tresor de la langue francaise, that the metalinguistic definition is indeed a category of the definitions but that, when compared to the other categories, it requires a different criteria of analysi, due to its nature. In order to do this, I shall present, first, the different nature of this issue from a typological approach on one side and a metalinguistic approach on the other. I shall expose, then, the main typological studies-in particular the unpublished document which is stored in the archives of the Laboratory ATILF [.Pour un nouveau cahier de normes...., 1979] as well as Martin (1983) and Rey-Debove (1998)-in which the question of the metalanguage is dealt with inside and following the example of typology to demonstrate that, if a definition such as aiguillette-nom populaire de l.orphie-is metalinguistic and a definition such as chaise - siege a dossier sans bras - is perifrastic, nom et siege are both hyperonyms, so that the typological criteria are not enough to distinguish between mealinguistic and perifrastic definition. Thus, I will establish, in accordance with Rey-Debove (1997), in which the definition is considered from a metalinguistic point of view-according to the sintactic relation between a lexical entry and its lexicographical definition, the principles which govern the metalinguistic analysis. The results will lead to three different categories of metalinguistic definitions of the syncategorematic words: 1. the definition refers to both infralinguistic and extralinguistic reality-in this case two sub-categories are possible: a) the hyperonym refers to the infralinguistic reality while the specific semes refer to the extralinguistic reality; b) the hyperonym refers to the infralinguistic reality while the specific semes, among which there is at least an autonym with 'schize' (cf. Rey-Debove 1997: 116-118), refer to the extralinguistic reality; 2. the definition refers to the only infralinguistic reality; 3. the definition refers to the only extralinguistic reality.
Studies examining factors that influence when words are learned typically investigate one lexical category or a small set of words. We provide the first evaluation of the relation between input frequency and age of acquisition for a large sample of words. The MacArthur-Bates Communicative Development Inventory provides norming data on age of acquisition for 562 individual words collected from the parents of children aged 0; 8 to 2; 6. The CHILDES database provides estimates of frequency with which parents use these words with their children (age: 0; 7-7; 5; mean age: 36 months). For production, across all words higher parental frequency is associated with later acquisition. Within lexical categories, however, higher frequency is related to earlier acquisition. For comprehension, parental frequency correlates significantly with the age of acquisition only for common nouns. Frequency effects change with development. Thus, frequency impacts vocabulary acquisition in a complex interaction with category, modality and developmental stage.
Abstract: The Web 2.0 maximizes Internet concept of encouraging its users to cooperate effectively for offer of virtual services and content organization. Among various potentialities of Web 2.0, folksonomy appears as a result of free attribution of tags to Web's resources by user himself. Folksonomies describe Web's resources; however, they aren't integrated in metadata in general. In order for them to be intelligible by machines and therefore used in Semantics Web context, they have to be automatically allocated to specific metadata elements. There are many metadata patterns. The focus of this investigation will be Dublin Core (DC) which is a gathering of metadata for description of electronic resources and which has been adopted by Institutional Repositories as a way of standardization and interoperability. We propose an investigation which intends to identify of metadata originated from folksonomies and integrate them in a DC Ontology extended so as to allow that values reported in tags may be conveniently gathered by protocol for metadata harvesting, specifically Open Archives Initiative - Protocol for Metadata Harvesting (OAI-PMH). This paper will present results of pilot study developed in beginning of investigation as well as metadata preliminarily defined. Metadata may be defined as a group of for description of resources [1]. There are many standards of metadata, however, in repository context; we can point out Dublin Core Metadata Element Set (DCMES) or simply Dublin Core (DC) which is a metadata pattern for description of electronic resources. This standard is well diffused, used globally and on a broad scale due to some factors: a) it was created specifically for description of electronic elements; b) it has an initiative which is responsible for its development, maintenance and spreading, Dublin Core Metadata Initiative (DCMI); c) it is group of metadata used for protocol Open Archives Initiative - Protocol for Metadata Harvesting (OAI-PMH), a mechanism for data transfer between digital repositories. The insertion of metadata in repositories may be done by authors themselves, professionals who mediate deposit or of final users. The more active participation of users in construction and organization of Internet contents is result of evolution of technologies used in Web, so-called Web 2.0, it is 'the network as platform, spanning all connected devices; Web 2.0 applications are those that make most of intrinsic advantages of that platform: delivering software as a continually-updated service that gets better more people use it, consuming and remixing data from multiple sources, including individual users, while providing their own data and services in a form that allows remixing by others, creating network effects through an architecture of participation, and going beyond page metaphor of Web 1.0 to deliver rich user experiences.' [2]. Among possibilities of Web 2.0 Folksonomy comes up as the result of personal free tagging of and objects (anything with an URL) for one's own retrieval. The tagging is done in a social environment (shared and open to others). The act of tagging is done by person consuming information [3]. The tags which make up a folksonomy would be key-words, categories or metadata [4]. In this brief definition of tag, we can notice that when attributed by users they can represent different roles. In a study [5][6] following roles are pointed out: Identifying What (or Who) it is About, Identifying What it Is, Identifying Who Owns It, Refining Categories, Identifying Qualities or Characteristics, Self Reference and Task Organizing. In another study, Kinds of Tags (KoT), which compared tags with DC metadata elements, authors observed that there are some tags which cannot be inserted in any of already existing and therefore, concluded that other metadata may be defined in order to include descriptions arising from folksonomies. Some probable which were identified: Action_Towards_Resource, To_Be_Used_In, Rate e Depth [7][8]. The KoT is being developed in partnership with following universities: Universidade do Minho (Portugal), University of Bologna (Italy), UKOLN (United Kingdom), Universidad Carlos III (Spain), La Trobe University (Australia) and Universit? Libr? de Bruxelles -Facult? de Philosophie et Lettres (Belgium) and has objective of verifying how tags derived from folksonomies can be normalized aiming at their interoperability with metadata standards, specifically DC. Summing up, metadata are groups of for description of digital resources, holding different standards, among them DC which is adopted by Repositories as basis for protocol for metadata harvesting (o OAI-PMH). In Web 2.0 context, folksonomies arise, which are result of Web resource tagging by its own users. Tags are a complementary form of description which expresses user's view of resource being used. It can be observed through preliminary results of KoT project that current of description defined in DCMI Metadata Terms do not include all descriptive attributed by resource users by means of these tags. In context shown, giving continuity to analysis resulting from KoT project, we propose an investigation which aims at identifying metadata derived from folksonomies and integrate them in a DC Ontology extended so as to enable that values reported in tags may be conveniently gathered by protocol for metadata harvesting. Being so, we intend to develop a qualitative approach research and answer following questions: Q1 - Which metadata are necessary to contain folksonomy values?; Q2 - Which metadata should be created and which is their relation with already existing ones in Dublin Core?; Q3 - Which codification schemes should be used and what is their relation to already recommended by DCMI?; Q4 - Which Ontology related to DC already enable access to previously established conceptualizations? Q5 - Accomplishing what is stipulated in DCAM, what is extension of DC Ontology which should be made available openly? The procedures are divided in four stages: 1) Analysing tags contained in KoT project dataset- at this stage we will analyse all tags in relation to resources to which they have been attributed. Complementarily, to settle doubts, it will be necessary to turn to lexical resources (dictionaries, encyclopaedias, Word Net, Wikipedia, etc) and to analyse tags in relation to its users to understand functionality of tag attributed as a metadata element. At this stage a pilot study will be developed to refine methodology proposed to verify if variants proposed for grouping and analysing tags are adequate to identify probable new metadata elements which could be extracted from folksonomies. 2) Propose complementary metadata to DC - Establishing description originated from folksonomies based on DC standard, DCAM model, ISO Standard 15836-2003 and NISO Standard Z39.85-2007 norms. At this stage we intend to propose and/or qualifiers complementary to DC. 3) Forming an Ontology - Here we intend to fulfil Integration of DC Ontologies with and/or qualifiers derived from folksonomies. The ontology will be created from Prot?g? tool and coded in OWL. 4) Validation of proposal - carried out by scientific community as methodology and results obtained will be presented in relevant events and scientific magazines and by DCMI Social Tagging community through investigations via online questionnaires and workshops proposed to community. It is intended that results of research may provide support so that applications based on Artificial Intelligence permit automation of harvesting processes including description provided from folksonomies. This paper will present results of pilot study (that is being finalized) alongside with preliminary results of first research stage: tag analysis. This stage will be done in following phases: a) Analysis and grouping of tags in their variant forms; b) Analysis of tags in relation to DC metadata and its qualifiers. The preliminary results of KoT point to possible proposal of some metadata or element refinements to DCMI. Those terms will potentially accommodate tags that currently do not have a metadata holder. The results of this research will therefore allow to determinate if KoT preliminary findings are verified and in which extension. The final paper will conclude with this discussion.
We survey the evaluation methodology adopted in information extraction (IE), as defined in a few different efforts applying machine learning (ML) to IE. We identify a number of critical issues that hamper comparison of the results obtained by different researchers. Some of these issues are common to other NLP-related tasks: e.g., the difficulty of exactly identifying the effects on performance of the data (sample selection and sample size), of the domain theory (features selected), and of algorithm parameter settings. Some issues are specific to IE: how leniently to assess inexact identification of filler boundaries, the possibility of multiple fillers for a slot, and how the counting is performed. We argue that, when specifying an IE task, these issues should be explicitly addressed, and a number of methodological characteristics should be clearly defined. To empirically verify the practical impact of the issues mentioned above, we perform a survey of the results of different algorithms when applied to a few standard datasets. The survey shows a serious lack of consensus on these issues, which makes it difficult to draw firm conclusions on a comparative evaluation of the algorithms. Our aim is to elaborate a clear and detailed experimental methodology and propose it to the IE community. Widespread agreement on this proposal should lead to future IE comparative evaluations that are fair and reliable. To demonstrate the way the methodology is to be applied we have organized and run a comparative evaluation of ML-based IE systems (the Pascal Challenge on ML-based IE) where the principles described in this article are put into practice. In this article we describe the proposed methodology and its motivations. The Pascal evaluation is then described and its results presented.
This article examines the constitution of ‘Japanese women’s language,’ or joseego, as a prescriptive linguistic norm for women through an analysis of scholarly and media representations of linguistic femininity in contemporary Japan. We distinguish two kinds of interrelated norms as constituting linguistic femininity: norms centered around general stylistic features such as politeness, gentleness, and refinement (the first-order norms) and those specificying particular linguistic forms including phonological, morphological, and lexical features (the second-order norms). Our analysis shows that both scholarly and media representations tend to share the dominant ideology of feminine speech in terms of general stylistic features and that they both link those features to specific linguistic forms in terms of Standard Japanese, but that the media representations are relatively more flexible in this linkage than the former and allows more room for contestation and the negotiation of alternative femininities. Through this analysis, we discuss the complex indexical process in which linguistic forms are ideologically linked to femininity as well as the tenuous nature of this linkage.
While large-scale corpora and various corpus query tools have long been recognized as essential language resources, the value of word association norms as language resources has been largely overlooked. This paper conducts some initial comparisons of the lexical relationships observed within Japanese collocation data extracted from a large corpus using the Japanese language version of the Sketch Engine (SkE) tool (Srdanović et al., 2008) and the relationships found within Japanese word association sets taken from the large-scale Japanese Word Association Database (JWAD) under ongoing construction by Joyce (2005, 2007). The comparison results indicate that while some relationships are common to both linguistic resources, many lexical relationships are only observed in one resource. These findings suggest that both resources are necessary in order to more adequately cover the diverse range of lexical relationships. Finally, the paper reflects briefly on the implementation of association-based word-search strategies into electronic dictionaries proposed by Zock and Bilac (2004) and Zock (2006).
Spoken language resources (SLRs) are essential for both research and application development. In this article we clarify the concept of SLR validation. We define validation and how it differs from evaluation. Further, relevant principles of SLR validation are outlined. We argue that the best way to validate SLRs is to implement validation throughout SLR production and have it carried out by an external and experienced institute. We address which tasks should be carried out by the validation institute, and which not. Further, we list the basic issues that validation criteria for SLR should address. A standard validation protocol is shown, illustrating how validation can prove its value throughout the production phase in terms of pre-validation, full validation and pre-release validation.
This paper describes an ongoing effort to build a large-scale monolingual treebank of parallel/comparable Dutch text, where nodes of syntax trees are aligned and labeled according to a small set of semantic similarity relations.Such a corpus has many potential uses and applications in e.g., multi-document summarization, question-answering and paraphrase extraction.We describe the text material, preprocessing, annotation, and alignment of sentences and syntax trees, both manual and automatic.Two new annotation tools are presented, as well as results from pilot experiments on inter-annotator agreement.On the basis of this resource, new automatic alignment software and NLP applications will be developed.
ProPOSEL is a prototype prosody and PoS (part-of-speech) English lexicon for Language Engineering, derived from the following language resources: the computer-usable dictionary CUVPlus, the CELEX-2 database, the Carnegie-Mellon Pronouncing Dictionary, and the BNC, LOB and Penn Treebank PoS-tagged corpora. The lexicon is designed for the target application of prosodic phrase break prediction but is also relevant to other machine learning and language engineering tasks. It supplements the existing record structure for wordform entries in CUVPlus with syntactic annotations from rival PoS-tagging schemes, mapped to fields for default closed and open-class word categories and for lexical stress patterns representing the rhythmic structure of wordforms and interpreted as potential new text-based features for automatic phrase break classifiers. The current version of the lexicon comes as a textfile of 104052 separate entries and is intended for distribution with the Natural Language ToolKit; it is therefore accompanied by supporting Python software for manipulating the data so that it can be used for Natural Language Processing (NLP) and corpus-based research in speech synthesis and speech recognition. 1. ProPOSEL: Derivation and Rationale A pronunciation lexicon is an integral part of the front-end NLP module in a generic Text-to-Speech (TTS) synthesis system and constitutes a natural way of giving such a system both prosodic and syntactic insights into input text. For English, three such resources- originally developed for Automatic Speech Recognition (ASR) and listing words and their phonetic transcriptions- are widely used: CELEX-2 (Baayen et al, 1996); PRONLEX (Kingsbury et
THE VOICE TEACHER IS REGULARLY BESET WITH CHALLENGES in the studio regarding consonant clusters in sung German, as is the singer who approaches any vocal work in the German language. The reputation of the German language as being consonant rather than vowel oriented is commonly appreciated and justifiable. Statistical studies show that the burden of text intelligibility is carried principally by the consonants, to a greater extent than most languages. A language that can produce lexical items such as entsturzt [ent'∫tYrtst] and kraftstrotzend ['kraft∫trctsent] adopts a strongly marked position among the world's languages with respect to the involvement of consonants in its sound system. These words contain ten and fourteen phonemes respectively, of which only two or three are vowels. The remaining clusters of consonants are samples of the subject of this article. The consonant clusters normally encountered in German phonology will be inventoried and contrasted with English. The material is likely to be familiar to many readers, albeit presented in a different, perhaps more systematic perspective than is normally encountered. The subject of German consonant clusters is best dealt with in terms of phonetic, not orthographic consonants. A firm grasp of the relationship between spelling and pronunciation is naturally also essential. Two or three successive letters may represent a single phoneme, as in [arrow right] /c/ or /x/ [arrow right] /∫/ [arrow right] /k/ [arrow right] /t/ Conversely, a single written consonant may serve to indicate more than one phoneme, as in [arrow right] /ts/ This situation is familiar because it is even more pronounced in English. The word scythe contains two consonant digraphs and two letter-vowels, but phonetically only one diphthong and no clusters at all. Since the greatest challenge in consonant clusters is visual (i.e., orthographic), thinking in terms of phonetic consonants should serve to simplify the matter for a student. German, more than most other languages, has absorbed lexical items from other languages into its own vocabulary, particularly from English, French, and Italian. Thus Duden, the principal lexicographic publisher in modern Germany, devotes an entire book to Fremdworter in its series of dictionaries. The process of lexical transfer is a complex aspect of German linguistics, particularly regarding pronunciation norms. Some words, such as Situation, have been subsumed into the phonological patterning of German, while others have retained the pronunciation of the word in the language from whence it came, or have struck a middle ground, such as Orange and Weekend. This diversity of phonetic transfer gives modern spoken German a particular flavor, and reflects the country's central geographic position in Europe. There are similar examples in English, such as cul-de-sac (where the French pronunciation has been distorted) and naive (which retains the original, although English idiosyncratically employs only the feminine form). This article will confine itself to the consonant clusters that occur regularly in the standard lexis, referring to combinations resulting from foreign influences only when appropriate. It is useful to consider consonant clusters in two quite distinct groups: syllable-interior, and across syllable or word boundaries. Part I of the article will concern itself with the former; Part II (to appear in the March/April 2008 issue), with the latter. Recognition of which group an example belongs to is the first step toward establishing correct pronunciation, and in some cases is necessary to discriminate between two potentially correct pronunciations. Before outlining in tabular form the cluster environments of German and English, it will be useful to consider all the consonantal combinations that are admissible in each language. A detailed theoretical account of the phonotactic rules and constraints of each language will not be necessary for our purposes. …
Since norms for vocabulary acquisition in Maltese children do not yet exist, documentation of productive vocabulary acquisition may contribute to establishing a baseline of lexical development. Clinical implications may thus be derived. The current study is a small-scale investigation of the proportions of Maltese and English lexemes in the vocabularies of ten normally-developing Maltese children aged between 12 and 30 months. The participants were primarily exposed to Maltese within their immediate environments, while receiving indirect exposure to English. Outcomes of parental report and language sampling were analysed for evidence of a bilingual dimension in these children's productive vocabularies. Translation equivalents were reported on by parents, but negligible evidence of equivalents emerged in conversational language use. In contrast, lexical borrowings were both reported and sampled. A substantial proportion of English lexemes were reported by the parents in the absence of Maltese equivalents.
Correct identification of word meaning is a long-standing problem for lexicography, language teaching, linguistic theory, and computer processing of text. Traditional approaches typically proceed word by word, relying on evidence from introspection – and have failed. A new theory of meaning is needed. In this prototype-based approach, called the Theory of Norms and Exploitations (TNE), the first step is identifying the phraseological patterns with which each word is associated. Meanings are then associated with patterns, rather than with isolated words. Words are highly ambiguous, but patterns are mostly unambiguous. \n Patterns cannot be identified by valency alone, but require statistical analysis and semantic typing of collocates. For example, (1) blowing up a bridge and (2) blowing up a balloon activate different meanings of blow up. But how many other contexts have the same effect on the meaning of the phrasal verb? Relevant members of the lexical set for (1) include building, factory, house, hotel, etc. Such lexical sets provide a basis for machine learning and text processing. \n Authentic uses of words are classified either as normal components of a pattern or as exploitations of norms. For example, “blowing up a condom” is not normal, but exploits (2). Creative metaphors are also exploitations.
Our study of seven first‐year college‐level German textbooks surveys how these texts deal with the challenge of representing gender‐inclusive language. Specifically, we look at gender marking in human nouns to see whether they occur as single‐gender forms (e.g., Student ), as morphological pairs in either full form (e.g., Student/Studentin ) or orthographically‐reduced form ( Student/in, StudentIn ), as lexical pairs (e.g., Großmutter/Großvater ) or as gender‐neutral forms (e.g., Studierende, plural ). All textbooks surveyed show a strong tendency to avoid gender‐exclusive forms but none introduces students to orthographically‐reduced pairs.