Towards the automatic merging of language resources

Language Resources are a critical component for Natural Language Processing applications. Throughout the years many resources were manually created for the same task, but with different granularity and coverage information. To create richer resources for a broad range of potential reuses, nformation from all resources has to be joined into one. The hight cost of comparing and merging different resources by hand has been a bottleneck for merging existing resources. With the objective of reducing human intervention, we present a new method for automating merging resources. We have addressed the merging of two verbs subcategorization frame (SCF) lexica for Spanish. The results achieved, a new lexicon with enriched information and conflicting information signalled, reinforce our idea that this approach can be applied for other task of NLP.

Towards a User-Friendly Platform for Building Language Resources based on Web Services

This paper presents the platform developed in the PANACEA project, a distributed factory that automates the stages involved in the acquisition, production, updating and maintenance of Language Resources required by Machine Translation and other Language Technologies. We adopt a set of tools that have been successfully used in the Bioinformatics field, they are adapted to the needs of our field and used to deploy web services, which can be combined to build more complex processing chains (workflows). This paper describes the platform and its different components (web services, registry, workflows, social network and interoperability). We demonstrate the scalability of the platform by carrying out a set of massive data experiments. Finally, a validation of the platform across a set of required criteria proves its usability for different types of users (non-technical users and providers).

From a pattern-language to a pattern ontology approach for CSCL script design

Collaborative activities, in which students actively interact with each other, have proved to provide significant learning benefits. In Computer-Supported Collaborative Learning (CSCL), these collaborative activities are assisted by technologies. However, the use of computers does not guarantee collaboration, as free collaboration does not necessary lead to fruitful learning. Therefore, practitioners need to design CSCL scripts that structure the collaborative settings so that they promote learning. However, not all teachers have the technical and pedagogical background needed to design such scripts. With the aim of assisting teachers in designing effective CSCL scripts, we propose a model to support the selection of reusable good practices (formulated as patterns) so that they can be used as a starting point for their own designs. This model is based on a pattern ontology that computationally represents the knowledge captured on a pattern language for the design of CSCL scripts. A preliminary evaluation of the proposed approach is provided with two examples based on a set of meaningful interrelated patters computationally represented with the pattern ontology, and a paper prototyping experience carried out with two teaches. The results offer interesting insights towards the implementation of the pattern ontology in software tools.

The development of intercultural awareness and changes of beliefs: the effects of studying abroad on learners of English as a second language

The present study examines the development of interculturality and changes of beliefs, by analyzing 106 compositions produced by 53 advanced level university students of translation studies at a university in Spain before and shortly after a stay-abroad (SA) period. The study draws on data collected at two different times: before (T1) and after the SA (T3). In addition, we compared the results with the writings produced by a control group of 10 native English speakers on SA too. Data were collected by means of a composition which tried to elicit the learners’ opinion about cultural habits maintenance. The results reveal significant changes between T1 and T3 in the degree of better attitudes and intercultural acquisition.

8-month non-use of a language. A case tudy. A longitudinal study of the effects on non use of one of the languages (Spanish) of a German L1 multilingual

Under the Dynamic Model of Multilingualism multilinguals are especially vulnerable to language attrition. It was the aim of the present study to verify if this was the case and to observe whether the different linguistic skills (receptive vs. descriptive) and the different linguistic levels (syntactic, lexical, morphological, etc.) would be affected equally.Data were gathered longitudinally by means of a language test for the subject’s reading, writing, listening and speaking skills as well as her knowledge of grammar and vocabulary. Although the overall accuracy remained intact and no proof for attrition in the receptive skills was found, the productive skills - mainly fluency - were shown to have suffered from language attrition. This was demonstrated by an increase in the number of pauses, hesitations, repetitions and self-corrections among others and decrease in the percentage of error-free clauses and decrease in the clause length, in oral and written fluency respectively.

Late bloomer or language impaired? Screening the speech of a Spanish-English bilingual toddler for early signs of language impairment

This case study presents corpus data gathered from a Spanish-English bilingual child with expressive language delay. Longitudinal data on the child’s linguistic development was collected from the onset of productive speech at age 1;1 until age 4 over the course of 28 video-taped sessions with the child’s principal caregivers. A literature review focused on the relationship between language delay and persisting disorders—including a discussion of the frequent difficulty in distinguishing between the two at early stages of bilingual development—is followed by an analysis of the child’s productive development in 2 distinct phases. An attempt is made to assess the child’s speech at age 4 for preliminary signs of SLI and to consider techniques for identifying ‘at risk’ bilingual children (that is, those with productive language delay, poor oral fluency, and family history of language problems) based on samples of recorded and transcribed speech.

Language acquisition during a stay abroad period following formal instruction: temporal effects on oral fluency development

This study investigates the development of fluency in 30 advanced L2 learners of English over a period of 15 months. In order to measure fluency, several temporal variables and hesitation phenomena are analyzed and compared. Oral competence is assessed by means of an oral interview carried out by the learners. Data collection takes place at three different times: before (T1) and after (T2) a six-month period of FI (80 hours) in the home university, and after a three-month SA term (T3). The data is analyzed quantitatively. Developmental gains in fluency are measured for the whole period, adopting a view of complementarity between the two learning contexts. From these results, a group of high fluency speakers is identified. Correlations between fluency gains and individual and contextual variables are executed and a more qualitative analysis is performed for high fluency speakers' performance and behavior. Results show an overall development of students' oral fluency during a period of 15 months favored by the combination of a period of FI at home followed by a 3-months SA.

Testing I(1) against I(d) alternatives in the presence of deteministic components

This paper discusses the role of deterministic components in the DGP and in the auxiliary regression model which underlies the implementation of the Fractional Dickey-Fuller (FDF) test for I(1) against I(d) processes with d ∈ [0, 1). This is an important test in many economic applications because I(d) processess with d & 1 are mean-reverting although, when 0.5 ≤ d & 1,, like I(1) processes, they are nonstationary. We show how simple is the implementation of the FDF in these situations, and argue that it has better properties than LM tests. A simple testing strategy entailing only asymptotically normally distributed tests is also proposed. Finally, an empirical application is provided where the FDF test allowing for deterministic components is used to test for long-memory in the per capita GDP of several OECD countries, an issue that has important consequences to discriminate between growth theories, and on which there is some controversy.

Panel index VAR models: Specification, Estimation, Testing and Leading Indicators

This paper proposes a method to conduct inference in panel VAR models with cross unit interdependencies and time variations in the coefficients. The approach can be used to obtain multi-unit forecasts and leading indicators and to conduct policy analysis in a multiunit setups. The framework of analysis is Bayesian and MCMC methods are used to estimate the posterior distribution of the features of interest. The model is reparametrized to resemble an observable index model and specification searches are discussed. As an example, we construct leading indicators for inflation and GDP growth in the Euro area using G-7 information.

Stepwise multiple testing as formalized data snooping

It is common in econometric applications that several hypothesis tests arecarried out at the same time. The problem then becomes how to decide whichhypotheses to reject, accounting for the multitude of tests. In this paper,we suggest a stepwise multiple testing procedure which asymptoticallycontrols the familywise error rate at a desired level. Compared to relatedsingle-step methods, our procedure is more powerful in the sense that itoften will reject more false hypotheses. In addition, we advocate the useof studentization when it is feasible. Unlike some stepwise methods, ourmethod implicitly captures the joint dependence structure of the teststatistics, which results in increased ability to detect alternativehypotheses. We prove our method asymptotically controls the familywise errorrate under minimal assumptions. We present our methodology in the context ofcomparing several strategies to a common benchmark and deciding whichstrategies actually beat the benchmark. However, our ideas can easily beextended and/or modied to other contexts, such as making inference for theindividual regression coecients in a multiple regression framework. Somesimulation studies show the improvements of our methods over previous proposals. We also provide an application to a set of real data.

Global Nash convergence of Foster and Young's regret testing

We construct an uncoupled randomized strategy of repeated play such that, if every player follows such a strategy, then the joint mixed strategy profiles converge, almost surely, to a Nash equilibrium of the one-shot game. The procedure requires very little in terms of players' information about the game. In fact, players' actions are based only on their own past payoffs and, in a variant of the strategy, players need not even know that their payoffs are determined through other players' actions. The procedure works for general finite games and is based on appropriate modifications of a simple stochastic learningrule introduced by Foster and Young.

Testing I(1) against I(d) alternatives in the presence of deteministic components

This paper discusses the role of deterministic components in the DGP and in the auxiliaryregression model which underlies the implementation of the Fractional Dickey-Fuller (FDF) test for I(1) against I(d) processes with d [0, 1). This is an important test in many economic applications because I(d) processess with d < 1 are mean-reverting although, when 0.5 = d < 1, like I(1) processes, they are nonstationary. We show how simple is the implementation of the FDF in these situations, and argue that it has better properties than LM tests. A simple testing strategy entailing only asymptotically normally distributedtests is also proposed. Finally, an empirical application is provided where the FDF test allowing for deterministic components is used to test for long-memory in the per capita GDP of several OECD countries, an issue that has important consequences to discriminate between growth theories, and on which there is some controversy.

Bringing game theory to hypothesis testing: Establishing finite sample bounds on inference

Small sample properties are of fundamental interest when only limited data is avail-able. Exact inference is limited by constraints imposed by speci.c nonrandomizedtests and of course also by lack of more data. These e¤ects can be separated as we propose to evaluate a test by comparing its type II error to the minimal type II error among all tests for the given sample. Game theory is used to establish this minimal type II error, the associated randomized test is characterized as part of a Nash equilibrium of a .ctitious game against nature.We use this method to investigate sequential tests for the di¤erence between twomeans when outcomes are constrained to belong to a given bounded set. Tests ofinequality and of noninferiority are included. We .nd that inference in terms oftype II error based on a balanced sample cannot be improved by sequential sampling or even by observing counter factual evidence providing there is a reasonable gap between the hypotheses.

Testing for convergence clubs in income per-capita: A predictive density approach

The paper proposes a technique to jointly test for groupings of unknown size in the cross sectional dimension of a panel and estimates the parameters of each group, and applies it to identifying convergence clubs in income per-capita. The approach uses the predictive density of the data, conditional on the parameters of the model. The steady state distribution of European regional data clusters around four poles of attraction with different economic features. The distribution of incomeper-capita of OECD countries has two poles of attraction and each grouphas clearly identifiable economic characteristics.

Testing the descriptive performance of the rank-dependent utility in the domain of health profiles

Expected utility theory (EUT) has been challenged as a descriptive theoryin many contexts. The medical decision analysis context is not an exception.Several researchers have suggested that rank dependent utility theory (RDUT)may accurately describe how people evaluate alternative medical treatments.Recent research in this domain has addressed a relevant feature of RDU models-probability weighting-but to date no direct test of this theoryhas been made. This paper provides a test of the main axiomatic differencebetween EUT and RDUT when health profiles are used as outcomes of riskytreatments. Overall, EU best described the data. However, evidence on theediting and cancellation operation hypothesized in Prospect Theory andCumulative Prospect Theory was apparent in our study. we found that RDUoutperformed EU in the presentation of the risky treatment pairs in whichthe common outcome was not obvious. The influence of framing effects onthe performance of RDU and their importance as a topic for future researchis discussed.