Search

[EN] The application of the latest Natural Language Processing breakthroughs in computational argumentation has shown promising results, which have raised the interest in this area of research. However, the available corpora with argumentative annotations are often limited to a very specific purpose or are not of adequate size to take advantage of state-of-the-art deep learning techniques (e.g., deep neural networks). In this paper, we present VivesDebate, a large, richly annotated and versatile professional debate corpus for computational argumentation research. The corpus has been created from 29 transcripts of a debate tournament in Catalan and has been machine-translated into Spanish and English. The annotation contains argumentative propositions, argumentative relations, debate interactions and professional evaluations of the arguments and argumentation. The presented corpus can be useful for research on a heterogeneous set of computational argumentation underlying tasks such as Argument Mining, Argument Analysis, Argument Evaluation or Argument Generation, among others. All this makes VivesDebate a valuable resource for computational argumentation research within the context of massive corpora aimed at Natural Language Processing tasks., This research was funded by the Spanish Government project PID2020-113416RB-I00; the Valencian Government project PROMETEO/2018/002; the MISMIS-Language project (PGC2018096212-B-C33) funded by Ministerio de Ciencia, Innovacion y Universidades; and the CLiC Research Group (2017SGR341) funded by Generalitat de Catalunya.

This article describes the criteria for identifying the focus of negation in Spanish. This work involved an in-depth linguistic analysis of the focus of negation through which we identified some 10 different types of criteria that account for a wide variety of constructions containing negation. These criteria account for all the cases that appear in the NewsCom corpus and were assessed in the annotation of this corpus. The NewsCom corpus consists of 2955 comments posted in response to 18 different news articles from online newspapers. The NewsCom corpus contains 2965 negative structures with their corresponding negation marker, scope, and focus. This is the first corpus annotated with focus in Spanish and it is freely available. It is a valuable resource that can be used both for the training and evaluation of systems that aim to automatically detect the scope and focus of negation and for the linguistic analysis of negation grounded in real data.

Research on speech technologies necessitates spoken data, which is usually obtained through read recorded speech, and specifically adapted to the research needs. When the aim is to deal with the prosody involved in speech, the available data must reflect natural and conversational speech, which is usually costly and difficult to get. This paper presents a machine learning-oriented toolkit for collecting, handling, and visualization of speech data, using prosodic heuristic. We present two corpora resulting from these methodologies: PANTED corpus, containing 250 h of English speech from TED Talks, and Heroes corpus containing 8 h of parallel English and Spanish movie speech. We demonstrate their use in two deep learning-based applications: punctuation restoration and machine translation. The presented corpora are freely available to the research community. This work was largely done during the doctoral studies of the first author in Universitat Pompeu Fabra. The second author has been funded by the Agencia Estatal de Investigación (AEI), Ministerio de Ciencia, Innovación y Universidades and the Fondo Social Europeo (FSE) under Grant RYC-2015-17239 (AEI/FSE, UE).

Code-switching is the phenomenon by which bilingual speakers switch between multiple languages during written or spoken communication. The importance of developing language technologies that are able to process code-switched language is immense, given the large populations that routinely code-switch. Current NLP and Speech models break down when used on code-switched data, interrupting the language processing pipeline in back-end systems and forcing users to communicate in ways which for them are unnatural. There are four main challenges that arise in building code-switched models: lack of code-switched data on which to train generative language models; lack of multilingual language annotations on code-switched examples which are needed to train supervised models; little understanding of how to leverage monolingual and parallel resources to build better code-switched models; and finally, how to use these models to learn why and when code-switching happens across language pairs. In this thesis, I look into different aspects of these four challenges. The first part of this thesis focuses on how to obtain reliable corpora of code-switched language. We collected a large corpus of code-switched language from social media using a combination of sets of anchor words that exist in one language and sentence-level language taggers. The newly obtained corpus is superior to other corpora collected via different strategies when it comes to the amount and type of bilingualism in it. It also helps train better language tagging models. We also have proposed a new annotation scheme to obtain part-of-speech tags for code-switched English-Spanish language. The annotation scheme is composed of three different subtasks including automatic labeling, word-specific questions labeling and question-tree word labeling. The part-of-speech labels obtained for the Miami Bangor corpus of English-Spanish conversational speech show very high agreement and accuracy. The second section of this thesis focuses on the tasks of part-of-speech tagging and language modeling. For the first task, we proposed a state-of-the-art approach to part-of-speech tagging of code-switched English-Spanish data based on recurrent neural networks.Our models were tested on the Miami Bangor corpus on the task of POS tagging alone, for which we achieved 96.34% accuracy, and joint part-of-speech and language ID tagging,which achieved similar POS tagging accuracy (96.39%) and very high language ID accuracy (98.78%). For the task of language modeling, we first conducted an exhaustive analysis of the relationship between cognate words and code-switching. We then proposed a set of cognate-based features that helped improve language modeling performance by 12% relative points. Furthermore, we showed that these features can also be used across language pairs and still obtain performance improvements. Finally, we tackled the question of how to use monolingual resources for code-switching models by pre-training state-of-the-art cross-lingual language models on large monolingual corpora and fine-tuning them on the tasks of language modeling and word-level language tagging on code-switched data. We obtained state-of-the-art results on both tasks.

In NLG systems, temporal uncertainty in raw data can hamper the inference of temporal and causal relationships between events and thus impact the quality of the generated texts. In this paper, we introduce a framework to represent and reason with temporal uncertainty based on possibility theory and propose a model that uses the outcomes of such temporal reasoning to select linguistic expressions to convey uncertainty to the reader. Our model is based on Fuzzy Temporal Constraint Networks (FTCN) and our work is based on the assumption that uncertainty should be communicated to an end user. The model we propose is grounded in experimental data from three languages. We present a large-scale empirical study that investigates the conditions that influence human subjective uncertainty in reasoning about temporal relations. Based on this, we also construct a classifier to select expressions to convey uncertainty, based on possibility and necessity values. We then present an evaluation which shows that the predictions of the FTCN model correlate well with human subjective uncertainty in different scenarios. An evaluation of our temporal expressions classifier also suggests good results, compared to human selection of linguistic expressions, as compared to baseline models., peer-reviewed

We present a novel heuristic approach that defines fuzzy geographical descriptors using data gathered from a survey with human subjects. The participants were asked to provide graphical interpretations of the descriptors ‘north’ and ‘south’ for the Galician region (Spain). Based on these interpretations, our approach builds fuzzy descriptors that are able to compute membership degrees for geographical locations. We evaluated our approach in terms of efficiency and precision. The fuzzy descriptors are meant to be used as the cornerstones of a geographical referring expression generation algorithm that is able to linguistically characterize geographical locations and regions. This work is also part of a general research effort that intends to establish a methodology which reunites the empirical studies traditionally practiced in data-to-text and the use of fuzzy sets to model imprecision and vagueness in words and expressions for text generation purposes., N/A

In neural image captioning systems, a recurrent neural network (RNN) is typically viewed as the primary `generation' component. This view suggests that the image features should be `injected' into the RNN. This is in fact the dominant view in the literature. Alternatively, the RNN can instead be viewed as only encoding the previously generated words. This view suggests that the RNN should only be used to encode linguistic features and that only the final representation should be `merged' with the image features at a later stage. This paper compares these two architectures. We find that, in general, late merging outperforms injection, suggesting that RNNs are better viewed as encoders, rather than generators., Comment: Appears in: Proceedings of the 10th International Conference on Natural Language Generation (INLG'17)

Maltese noun phrases exhibit a form of ‘definiteness agreement’ between head noun and modifier. When the noun is definite, an adjectival modifier is often overtly marked as definite as well. However, the status of this phenomenon as a case of true morphosyntactic agreement has been disputed, given its apparent optionality. Not all definite nps have modifiers which are overtly marked as definite. Some authors have argued that definiteness marking on the adjective is in fact pragmatically licensed. The present paper presents a corpus-based study of the distribution of adjectives with and without definite marking, and then tests the pragmatic licensing claim through a production study. Speakers were found to be more likely to use definite adjectives in referential noun phrases when the adjectives had a specifically contrastive function. This result is discussed in the context of both theoretical and psycholinguistic work on the pragmatics of referentiality., peer-reviewed

When producing a description of a target referent in a visual context, speakers need to choose a set of properties that distinguish it from its distractors. Computational models of language production/generation usually model this as a search process and predict that the time taken will increase both with the number of distractors in a scene and with the number of properties required to distinguish the target. These predictions are reminiscent of classic ndings in visual search; however, unlike models of reference production, visual search models also predict that search can become very e cient under certain conditions, something that reference production models do not consider. This paper investigates the predictions of these models empirically. In two experiments, we show that the time taken to plan a referring expression { as re ected by speech onset latencies { is in uenced by distractor set size and by the number of properties required, but this crucially depends on the discriminability of the properties under consideration. We discuss the implications for current models of reference production and recent work on the role of salience in visual search., peer-reviewed

Maltese is a morphologically rich language with a hybrid morphological system which features both concatenative and non-concatenative processes. This paper analyses the impact of this hybridity on the performance of machine learning techniques for morphological labelling and clustering. In particular, we analyse a dataset of morphologically related word clusters to evaluate the difference in results for concatenative and non-concatenative clusters. We also describe research carried out in morphological labelling, with a particular focus on the verb category. Two evaluations were carried out, one using an unseen dataset, and another one using a gold standard dataset which was manually labelled. The gold standard dataset was split into concatenative and non-concatenative to analyse the difference in results between the two morphological systems., non peer-reviewed

This article presents the Spanish Iarg-AnCora corpus (400 k-words, 13,883 sentences) annotated with the implicit arguments of deverbal nominalizations (18,397 occurrences). We describe the methodology used to create it, focusing on the annotation scheme and criteria adopted. The corpus was manually annotated and an interannotator agreement test was conducted (81 % observed agreement) in order to ensure the reliability of the final resource. The annotation of implicit arguments results in an important gain in argument and thematic role coverage (128 % on average). It is the first corpus annotated with implicit arguments for the Spanish language with a wide coverage that is freely available. This corpus can subsequently be used by machine learning-based semantic role labeling systems, and for the linguistic analysis of implicit arguments grounded on real data. Semantic analyzers are essential components of current language technology applications, which need to obtain a deeper understanding of the text in order to make inferences at the highest level to obtain qualitative improvements in the results.

This paper deals with the MUMIN multimodal annotation scheme, which is dedicated to the study of hand gestures and facial displays in interpersonal communication, with focus on the role played by multimodal expressions for feedback, turn management and sequencing. The scheme has been tested on the analysis of multimodal behaviour in short video clips in Swedish, Finnish and Danish. These preliminary results show that the categories defined are reliable, and points at a few necessary revisions., peer-reviewed

In this article we present the most relevant new aspects from the COD2 project, which is based on an oral dialectal corpus of contemporary Catalan created as a continuation of the first COD compiled two decades before. In the first place, the new project intends to analyze the changes undergone by the Catalan language during this period. In second place, we no longer center the linguistic analysis on the traditional generativist techniques but on the tenets of Optimality Theory. And, finally, we use new dialectometric techniques based principally on probabilistic analytical procedures.

This response discusses the experiment reported in Krahmer et al.’s Letter to the Editor of Cognitive Science. We observe that their results do not tell us whether the Incremental Algorithm is better or worse than its competitors, and we speculate about implications for reference in complex domains, and for learning from ‘‘normal’’ (i.e., non-semantically-balanced) corpora., peer-reviewed

This article introduces the topic ‘‘Production of Referring Expressions: Bridging the Gap between Computational and Empirical Approaches to Reference’’ of the journal Topics in Cognitive Science. We argue that computational and psycholinguistic approaches to reference production can benefit from closer interaction, and that this is likely to result in the construction of algorithms that differ markedly from the ones currently known in the computational literature. We focus particularly on determinism, the feature of existing algorithms that is perhaps most clearly at odds with psycholinguistic results, discussing how future algorithms might include non-determinism, and how new psycholinguistic experiments could inform the development of such algorithms., peer-reviewed

Until recently, referring expression generation (reg) research focused on the task of selecting the semantic content of definite mentions of listener-familiar discourse entities. In the grec research programme we have been interested in a version of the reg problem definition that is (i) grounded within discourse context, (ii) embedded within an ap- plication context, and (iii) informed by naturally occurring data. This paper provides an overview of our aims and motivations in this research programme, the data resources we have built, and the first three shared- task challenges, grec-msr’08, grec-msr’09 and grec-neg’09, we have run based on the data., peer-reviewed

Many real-world applications that reason about events obtained from raw data must deal with the problem of temporal uncertainty, which arises due to error or inaccuracy in data. Uncertainty also compromises reasoning where relationships between events need to be inferred. This paper discusses an approach to dealing with uncertainty in temporal and causal relations using Possibility Theory, focusing on a family of medical decision support systems that aim to generate textual summaries from raw patient data in a Neonatal Intensive Care Unit. We describe a framework to capture temporal uncertainty and to express it in generated texts by mean of linguistic modifiers. These modifiers have been chosen based on a human experiment testing the association between subjective certainty about a proposition and the participants’ way of verbalising it., peer-reviewed

This paper describes a study in which a corpus of spoken Danish annotated with focus and topic tags was used to investigate the relation between information structure and pauses. The results show that intra-clausal pauses in the focus domain, tend to precede those words that express the property or semantic type whereby the object in focus is distinguished from other ones in the domain., peer-reviewed

By providing a better understanding of paraphrase and coreference in terms of similarities and differences in their linguistic nature, this article delimits what the focus of paraphrase extraction and coreference resolution tasks should be, and to what extent they can help each other. We argue for the relevance of this discussion to Natural Language Processing.