id
stringlengths
20
52
title
stringlengths
3
459
abstract
stringlengths
0
12.3k
classification_labels
sequence
numerical_classification_labels
sequence
http://arxiv.org/abs/2207.11442v2
$μ\text{KG}$: A Library for Multi-source Knowledge Graph Embeddings and Applications
This paper presents $\mu\text{KG}$, an open-source Python library for representation learning over knowledge graphs. $\mu\text{KG}$ supports joint representation learning over multi-source knowledge graphs (and also a single knowledge graph), multiple deep learning libraries (PyTorch and TensorFlow2), multiple embedding tasks (link prediction, entity alignment, entity typing, and multi-source link prediction), and multiple parallel computing modes (multi-process and multi-GPU computing). It currently implements 26 popular knowledge graph embedding models and supports 16 benchmark datasets. $\mu\text{KG}$ provides advanced implementations of embedding techniques with simplified pipelines of different tasks. It also comes with high-quality documentation for ease of use. $\mu\text{KG}$ is more comprehensive than existing knowledge graph embedding libraries. It is useful for a thorough comparison and analysis of various embedding models and tasks. We show that the jointly learned embeddings can greatly help knowledge-powered downstream tasks, such as multi-hop knowledge graph question answering. We will stay abreast of the latest developments in the related fields and incorporate them into $\mu\text{KG}$.
[ "Semantic Text Processing", "Structured Data in NLP", "Representation Learning", "Knowledge Representation", "Multimodality" ]
[ 72, 50, 12, 18, 74 ]
http://arxiv.org/abs/1803.07771v1
$ρ$-hot Lexicon Embedding-based Two-level LSTM for Sentiment Analysis
Sentiment analysis is a key component in various text mining applications. Numerous sentiment classification techniques, including conventional and deep learning-based methods, have been proposed in the literature. In most existing methods, a high-quality training set is assumed to be given. Nevertheless, constructing a high-quality training set that consists of highly accurate labels is challenging in real applications. This difficulty stems from the fact that text samples usually contain complex sentiment representations, and their annotation is subjective. We address this challenge in this study by leveraging a new labeling strategy and utilizing a two-level long short-term memory network to construct a sentiment classifier. Lexical cues are useful for sentiment analysis, and they have been utilized in conventional studies. For example, polar and privative words play important roles in sentiment analysis. A new encoding strategy, that is, $\rho$-hot encoding, is proposed to alleviate the drawbacks of one-hot encoding and thus effectively incorporate useful lexical cues. We compile three Chinese data sets on the basis of our label strategy and proposed methodology. Experiments on the three data sets demonstrate that the proposed method outperforms state-of-the-art algorithms.
[ "Representation Learning", "Language Models", "Semantic Text Processing", "Sentiment Analysis" ]
[ 12, 52, 72, 78 ]
SCOPUS_ID:85067335226
“Should We Allow Him to Pass?” Increasing Cooperation Between Truck Drivers Using Anthropomorphism
Studies from various disciplines have showed, that adding human characteristics to non-human object improves the interaction between human and this object. It can be assumed that human-like technologies have a positive influence on driver-vehicle interaction as well. This study investigates the potential to increase the willingness of truck drivers to cooperate during overtaking scenarios using anthropomorphized interfaces. Therefore, a driving simulator experiment was conducted with truck drivers. Two conversational agents have been developed, which differ in their degree of human characteristics. They supported the truck driver in the initiation and during the overtaking manoeuver by clarifying a willingness to cooperate and communicating the status of the overtaking process. The results indicated no significance in the drivers’ willingness to cooperate in interaction with the two agents. However, the perceived human likeness increased through the addition of emotionality and identity. More than half of the drivers were in favor of the more human-like agent.
[ "Natural Language Interfaces", "Dialogue Systems & Conversational Agents" ]
[ 11, 38 ]
SCOPUS_ID:85149359254
<bold>A Comprehensive Review on Arabic Sarcasm Detection: Approaches, Challenges and Future Trends</bold>
On social media platforms, it is essential to express one’s thoughts, opinions, and reviews. Sarcasm is a widely used linguistic form to criticise or express a person’s ideas with ridicule where the written text has both an intended and unintended meaning. The sarcastic text frequently reverses the polarity of the sentiment. Therefore, detecting sarcasm has a positive impact on the sentiment analysis task and ensures more accurate results. Although Arabic is one of the most frequently used languages for web content sharing, the sarcasm detection of Arabic content is restricted and yet still naive due to several challenges, including the morphological structure of the Arabic language, the variety of dialects, and the lack of adequate data sources. In this paper, we review those studies on Arabic sarcasm detection that were published between 2017 and 2022 and identify the most prevalent Artificial Intelligence (AI) techniques, datasets, and challenges. In addition, we highlight the challenge of publicly available Arabic sarcasm datasets. We also look into the manual and automatic annotation methods used for corpora construction. Moreover, we highlight the most commonly used Arabic language models for sarcasm detection.
[ "Stylistic Analysis", "Sentiment Analysis" ]
[ 67, 78 ]
SCOPUS_ID:34248707753
<e2>Thomas D. Cravens</e2> (2002). <e1>Comparative historical dialectology: Italo-Romance clues to Ibero-Romance sound change</e1>. (Current Issues in Linguistic Theory 231.) Amsterdam & Philadelphia: John Benjamins Publishing Company. Pp. xi+163
[ "Linguistics & Cognitive NLP", "Linguistic Theories" ]
[ 48, 57 ]
SCOPUS_ID:85147210432
<inline-formula><tex-math notation="LaTeX">$D^{3}K$</tex-math></inline-formula>: Dynastic Data-Free Knowledge Distillation
Data-free knowledge distillation further broadens the applications of the distillation model. Nevertheless, the problem of providing diverse data with rich expression patterns needs to be further explored. In this paper, a novel dynastic data-free knowledge distillation (<inline-formula><tex-math notation="LaTeX">$D^{3}K$</tex-math></inline-formula>) model is proposed to alleviate this problem. In this model, a dynastic supernet generator (D-SG) with a flexible network structure is proposed to generate diverse data. The D-SG can adaptively alter architectural configurations and activate different subnet generators in different sequential iteration spaces. The variable network structure increases the complexity and capacity of the generator, and strengthens its ability to generate diversified data. In addition, a novel additive constraint based on the differentiable dhash (D-Dhash) is designed to guide the structure parameter selection of the D-SG. This constraint forces the D-SG to constantly jump out of the fixed generation mode and generate diverse data in semantics and instance. The effectiveness of the proposed model is verified on the experimental benchmark datasets (MNIST, CIFAR-10, CIFAR-100, and SVHN).
[ "Language Models", "Semantic Text Processing", "Responsible & Trustworthy NLP", "Reasoning", "Numerical Reasoning", "Green & Sustainable NLP" ]
[ 52, 72, 4, 8, 5, 68 ]
SCOPUS_ID:85147278864
&lt;sc&gt;AttSum&lt;/sc&gt;: A Deep Attention-Based Summarization Model for Bug Report Title Generation
Concise and precise bug report titles help software developers to capture the highlights of the bug report quickly. Unfortunately, it is common that bug reporters do not create high-quality bug report titles. Recent long short-term memory (LSTM)-based sequence-to-sequence models such as iTAPE were proposed to generate bug report titles automatically, but the text representation method and LSTM employed in such model are difficult to capture the accurate semantic information and draw the global dependencies among tokens effectively. This article proposes a deep attention-based summarization model (i.e., <sc>AttSum</sc>) to generate high-quality bug report titles. Specifically, the <sc>AttSum</sc> model employs the encoder.decoder framework, which utilizes the robustly optimized bidirectional-encoder-representations-from-transformers approach to encode the bug report bodies to capture contextual semantic information better, the stacked transformer decoder to automatically generate titles, and the copy mechanism to handle the rare token problem. To validate the effectiveness of <sc>AttSum</sc>, we conduct automatic and manual evaluations on 333563 &#x201C;<inline-formula><tex-math notation="LaTeX">$&lt; body, title&gt;$</tex-math></inline-formula>&#x201D; pairs of bug reports and perform a practical analysis of its ability to improve low-quality titles. The result shows that <sc>AttSum</sc> is superior to the state-of-the-art baselines by a substantial margin both on automatic evaluation metrics (e.g., by 3.4&#x0025;&#x2013;58.8&#x0025; and 7.7&#x0025;&#x2013;42.3&#x0025; in terms of recall-oriented understudy for gisting evaluation in F1 and bilingual evaluation understudy, separately) and three human-set modalities (e.g., by 1.9&#x0025;&#x2013;57.5&#x0025;). Moreover, we analyze the impact of the training data size on <sc>AttSum</sc> and the results imply that our approach is robust enough to generate much better titles.
[ "Language Models", "Semantic Text Processing", "Robustness in NLP", "Summarization", "Text Generation", "Responsible & Trustworthy NLP", "Reasoning", "Numerical Reasoning", "Information Extraction & Text Mining" ]
[ 52, 72, 58, 30, 47, 4, 8, 5, 3 ]
SCOPUS_ID:85144776716
&lt;sc&gt;InterEvo-TR&lt;/sc&gt;: Interactive Evolutionary Test Generation with Readability Assessment
Automated test case generation has proven to be useful to reduce the usually high expenses of software testing. However, several studies have also noted the skepticism of testers regarding the comprehension of generated test suites when compared to manually designed ones. This fact suggests that involving testers in the test generation process could be helpful to increase their acceptance of automatically-produced test suites. In this paper, we propose incorporating interactive readability assessments made by a tester into EvoSuite, a widely-known evolutionary test generation tool. Our approach, <sc>InterEvo-TR</sc>, interacts with the tester at different moments during the search and shows different test cases covering the same coverage target for their subjective evaluation. The design of such an interactive approach involves a schedule of interaction, a method to diversify the selected targets, a plan to save and handle the readability values, and some mechanisms to customize the level of engagement in the revision, among other aspects. To analyze the potential and practicability of our proposal, we conduct a controlled experiment in which 39 participants, including academics, professional developers, and student collaborators, interact with <sc>InterEvo-TR</sc>. Our results show that the strategy to select and present intermediate results is effective for the purpose of readability assessment. Furthermore, the participants&#x0027; actions and responses to a questionnaire allowed us to analyze the aspects influencing test code readability and the benefits and limitations of an interactive approach in the context of test case generation, paving the way for future developments based on interactivity.
[ "Semantic Text Processing", "Text Complexity" ]
[ 72, 42 ]
SCOPUS_ID:34248717421
'...And the dumb speak': George Lamming's theory of language and the epistemology of the body in The Emigrants
[ "Linguistics & Cognitive NLP", "Linguistic Theories" ]
[ 48, 57 ]
SCOPUS_ID:79952963718
'A beautiful show of strength': Weight loss and the fat activist self
This article explores the disciplinary and productive effects of late 20th-century/early 21st-century discourse around the obesity epidemic. For my purposes, both the Fat Activist and Weight Loss Surgery patient may be seen as outcomes of this discursive proliferation. Based on a content analysis of two online accounts of weight loss by prominent fat activists, I examine the ways in which such public declarations can be seen as contemporary examples of Foucault's notion of the confession. Through debates in the 'fatosophere', such activists and their peers simultaneously take up and reject both biomedical and fat activists tenets in ways that are highly ambivalent. In the process, fat activism unsettles and resituates its borders and boundaries and the issue of weight loss becomes a discursive technique of knowledge/power for the understanding of and relating to the fat activist self. © 2011 The Author(s).
[ "Discourse & Pragmatics", "Semantic Text Processing" ]
[ 71, 72 ]
SCOPUS_ID:84878770607
'A gypsy is a gypsy': A literary construction in a long-term perspective
This article analyses Swedish antiziganism (anti-Gypsyism) in a long-term perspective as a discursive formation deriving from the conceptual gypsy: a historically constructed image born of social, religious, and racial prejudice that is projected onto Roma and others. The source material has been chosen to reflect the intertextuality between the non-fictional gypsiology literature of the eighteenth and twentieth centuries: the influential work of the German historian Heinrich Moritz Grellmann is compared with the Swedish doctoral dissertations of Samuel P. Björckman (1730) and Laurentius Rabenius (1791); the three-part documentary Zigenare (1929,1955 and 1963) by the popular Proletarian School author Ivar Lo-Johansson; the civil rights intervention by Katarina Taikon from a Roma discourse position (Zigenare ar vi, 1963); and finally Karl-Olov Arnstberg's controversial study Svenskar och zigenare (1998).' The article posits a strong link between the academic antiziganism of the eighteenth and twentieth centuries, and the notions of Roma culture as essentially anti-social, and thus to be combated by forced assimilation; of Roma as ungrateful for the generosity of their hosts; and of antiziganism as caused by Roma behaviour. The continuity is obvious in the use of recurring reference codes, such as typologies of 'gypsy' crimes, used in a circular argument to slap the label of gypsy on actual and supposed Roma. The study also exposes the biological racism inherent in Swedish antiziganism, and in the case of Lo-Johansson clarifies the meaning of philoziganism as excluding and essentialist in its romantic projections onto Roma. Further, the article shows that 'zigenare' ('gypsy') in Swedish is unambiguously linked to a pejorative discourse, though in all the works discussed it is used in references to, but not always when referring to Roma, and is thus considered unsuitable, as well as unethical, as an analytical notion.
[ "Discourse & Pragmatics", "Semantic Text Processing" ]
[ 71, 72 ]
SCOPUS_ID:0031218038
'A way of struggle': Reformations and affirmations of E. P. Thompson's class analysis in the light of postmodern theories of language
This paper is an analysis of the role of language in historical class formation in light of the recent developments in postmodern social theory and historiography. Revisionists from within this perspective have questioned if not abandoned E. P. Thompson's class struggle analysis, arguing that he fails to account for the constitutive character of language in the construction of collective identities. They oppose his account of the making of the English working class with alternative histories emphasizing populist and other non-class identities. Drawing on the Bakhtin Circle of literary studies, and returning to Thompson's own writings, I argue that we can incorporate language into class struggle analysis as a critical mediating force. I maintain that class struggle occurs largely within a hegemonic discursive formation, and that class consciousness and identity thus in part are formed through counter-hegemonic strategies of resistance to ideological domination. To illustrate this theory I analyse the role of language in the class struggles of the silk weavers of the Spitalfields district in London in the 1820s. I analyse how the silk weavers articulated a class consciousness through their counter-hegemonic struggles with the large capitalists and the language of political economy.
[ "Linguistics & Cognitive NLP", "Linguistic Theories" ]
[ 48, 57 ]
SCOPUS_ID:85066849985
'A wisdom of crowds': Social media mining for soccer match analysis
The advent of social media has allowed channeling of the voice of sports fans that have essentially lead to gathering and storing fan-generated, large-scale opinions about sports match and team performance. Although research utilizing social media data for the purposes of supporting consumer market research has been increasing throughout the recent decade, there is a lack of studies using social media mining approach to improve team performance. In this paper, an opportunity mining approach is proposed to identify opportunities to improve team performance based on text mining and cluster analysis. A case study of the 2018 Fédération Internationale de Football Association (FIFA) World Cup final qualification of Korea, Korea versus Uzbekistan, was conducted to explain how the proposed method works. Fan comment data collected in the study revealed 16 different opportunities that would satisfy fans with regard to the team performance, and of those, two main extreme opportunities were identified.
[ "Information Extraction & Text Mining", "Text Clustering" ]
[ 3, 29 ]
SCOPUS_ID:0141993799
'About a year before the breakdown I was having symptoms': Sadness, pathology and the australian newspaper media
Portrayals of mental illness in the media reportedly highlight violence and crime by the 'mentally ill'. Using a discourse analytic approach we investigated representations of 'depression' in the print media in Australia during the year 2000. Unlike other 'mental illnesses', in the case of depression the media stress the need for the protection of the sufferer, rather than others. Three key discourses are identified - the biomedical, the psycho-social and the administrative/managerial - which work to normalise depression by presenting it as beyond the control of the afflicted individual: a consequence of faulty brain chemistry or the product of social conditions. These discourses work together to produce unhappiness as individualised pathology in need of management through biological, psychological or social structural controls.
[ "Discourse & Pragmatics", "Semantic Text Processing" ]
[ 71, 72 ]
SCOPUS_ID:84919433539
'Above the fray': Interests, discourse and legitimacy in the audit field
Legitimacy is a crucial concern for the institutional field of auditing, given its reliance on perceived legitimacy for its political mandate and license to practice, in addition to its wider credibility and trust amongst stakeholders such as clients, investors and the public. In this paper, we explore the role of interest-discourse in the discursive strategies of legitimization in the audit field. We develop an Ethnomethodologically informed Discourse Analysis (EDA) perspective that enables us to theorise how institutional actors account for interests as a means for de-legitimization and re-legitimization. We ask: how do institutional actors in the audit field establish who or what is 'above the fray' and who or what is 'interested'? We illustrate our argument by examining how the 'Big Four' audit firms handled a 'crisis of legitimacy' in the accountancy profession following the recent financial crisis, focussing in particular on a Parliamentary inquiry into market concentration in the audit industry in the UK. First, we show how de-legitimization is achieved through the discursive strategies of stake attribution, stake interrogation and stake mis-alignment. Second, we show how re-legitimization is achieved through the discursive strategies of stake inoculation, stake confession, stake alignment and stake transcendence. We conclude by examining the implications of the discursive processes we have studied for the future of the audit field in the UK.
[ "Discourse & Pragmatics", "Semantic Text Processing" ]
[ 71, 72 ]
SCOPUS_ID:84914162164
'Active play may be lots of fun, but it's certainly not frivolous': The emergence of active play as a health practice in Canadian public health
In the context of what has been termed a childhood obesity epidemic, public health institutions have recently begun to promote active play as a means of addressing childhood obesity, thus advancing play for health. Drawing on Foucault, this article problematises the way that children's play is being taken up as a health practice and further considers some of the effects this may have for children. Six Canadian public health websites were examined, from which 150 documents addressing children's health, physical activity, obesity, leisure activities and play were selected and coded deductively (theoretical themes) and inductively (emerging themes). Bacchi's question-posing approach to critical discourse analysis deepened our analysis of dominant narratives. Our findings suggest that several taken-for-granted assumptions and practices underlie this discourse: (i) play is viewed as a productive activity legitimises it as a health practice; (ii) tropes of 'fun' and 'pleasure' are drawn on to promote physical activity; (iii) children are encouraged to self-govern their leisure time to promote health. We underscore the need to recognise this discourse as contingent and as only one of many ways of conceptualising children's leisure activities and their health and social lives more generally.
[ "Discourse & Pragmatics", "Semantic Text Processing" ]
[ 71, 72 ]
SCOPUS_ID:84857882558
'Ageing well': Competing discourses and tensions in the management of knee pain
Age and ill-health have long been identifi ed as being intrinsically intertwined, however the boundary between the discourses of ageing and chronic illness is not easy to demarcate. It is possible, however, to shed some light on the interplay between the ageing body and chronic illness by considering how differing cultures of ageing play a key role in articulating a dichotomy between seeing chronic illness as an aspect of health or of ill-health and age-related decline. Using knee pain as an example we argue that to understand how the interplay between ageing and chronic illness infl uences the interpretation of knee pain as 'ageing well' or as a sign of impending decline and physical dependency, then we need to employ a discursive methodological approach which explores talk as a form of action designed for its local interactional context and pays attention to what statements mean in the context in which they occur. Copyright © eContent Management Pty Ltd.
[ "Discourse & Pragmatics", "Semantic Text Processing" ]
[ 71, 72 ]
SCOPUS_ID:85102257311
'Alien elections': neighboring state news on the 2018 russian presidential elections
News media tend to reflect voices in the political establishment while covering international events. Is it still true when almost half of the national audience speak the language of the country featured in the coverage? In this paper, we present an analysis of 19.5k news messages collected from Russian-language Ukrainian news outlets covering the 2018 presidential elections in Russia. Using a mixed-method approach (topic modeling and qualitative reading), we identify key topics and stories and evaluate the extent of personalization in the election coverage. We find three central angles: The focus on polls and election results, election preparations in Crimea, and Vladimir Putin's victory. The elections are linked predominantly to Crimean issues through the date of the elections, each candidate's stance on the subject, the election management in the region, and other countries' reactions to the results. Such coverage has an accusatory bias; it stresses the legal status of the Crimean referendum and the Russian authorities' actions and reports the pressures on locals by authorities, especially the Crimean Tatars. Not linked directly to Crimea, other angles are less emotionally charged. Political personalization of the discussion has a contradictory nature. On one hand, the overwhelming majority of the messages mention public figures. On the other hand, the coverage of the figures is limited and omits their traits. Moreover, at times, public figures are replaced by non-personalized symbols (e.g., Kremlin, Russian invaders). However, if the former's coverage is predominantly neutral, the latter's coverage is more prone to negative and loaded statements.
[ "Topic Modeling", "Information Extraction & Text Mining" ]
[ 9, 3 ]
SCOPUS_ID:85142918314
'Am I the Bad One'? Predicting the Moral Judgement of the Crowd Using Pre-trained Language Models
Natural language processing (NLP) has been shown to perform well in various tasks, such as answering questions, ascertaining natural language inference and anomaly detection. However, there are few NLP-related studies that touch upon the moral context conveyed in text. This paper studies whether state-of-the-art, pre-trained language models are capable of passing moral judgments on posts retrieved from a popular Reddit user board. Reddit is a social discussion website and forum where posts are promoted by users through a voting system. In this work, we construct a dataset that can be used for moral judgement tasks by collecting data from the AITA? (Am I the A*******?) subreddit. To model our task, we harnessed the power of pre-trained language models, including BERT, RoBERTa, RoBERTa-large, ALBERT and Longformer. We then fine-tuned these models and evaluated their ability to predict the correct verdict as judged by users for each post in the datasets. RoBERTa showed relative improvements across the three datasets, exhibiting a rate of 87% accuracy and a Matthews correlation coefficient (MCC) of 0.76, while the use of the Longformer model slightly improved the performance when used with longer sequences, achieving 87% accuracy and 0.77 MCC.
[ "Language Models", "Semantic Text Processing" ]
[ 52, 72 ]
SCOPUS_ID:84885466275
'Anchor baby': A conceptual explanation for pejoration
In this paper, I offer a detailed account of the pejorative nature of the term 'anchor baby', an increasingly common phrase used to frame the children of undocumented immigrants within the United States. Using cognitive linguistic methodology within the Critical Discourse Analysis paradigm [Chilton, Missing links in mainstream CDA: Modules, blends and the critical instinct. In Wodak, R. and Chilton, P. (Eds.) A New Agenda in (Critical) Discourse Analysis: Theory, Methodology and Interdisciplinarity. John Benjamins, Amsterdam, pp. 19-53, 2005], I explore language from both sides of the immigration debate to show, step by step, how a seemingly simple compound can subconsciously introduce or affirm deep value-laden judgments about both authorized and unauthorized immigrants. As such, my explanation rests on the theory of Conceptual Blending [G. Fauconnier, M. Turner, The Way We Think: Conceptual Blending and the Mind's Hidden Complexities. Basic Books, New York, 2002] in which multiple domains of knowledge including those of seafaring, immigration, and family structure blend together giving rise to a new (and offensive) concept of procreation as a tool to gain legal status, resources, and permanent ties to the U.S. I show that hidden meaning and reasoning, instinctively labeled as offensive by many, but often characterized in vague terms, can in fact be systematically described given the right conceptual framework. I propose the theory of Conceptual Blending can be utilized as a uniquely effective tool to dissect associative processes in the study of pragmatics. © 2013 Elsevier B.V.
[ "Semantic Text Processing", "Linguistic Theories", "Discourse & Pragmatics", "Explainability & Interpretability in NLP", "Ethical NLP", "Linguistics & Cognitive NLP", "Responsible & Trustworthy NLP" ]
[ 72, 57, 71, 81, 17, 48, 4 ]
SCOPUS_ID:80053040827
'Anita Musso' annotation according with Language into Act Theory
The paper analyses a brief stretch of a spoken French interview and presents its annotation according to the Language in Act Theory (transcription, alignment, informational functions). The text, as most non professional interviews, shares formal and informal features; i.e. even if spontaneous and unscripted, it lacks the strong pragmatic characters of every day talks. Systematic quantitative analysis of the text are presented (dialogic turns, words, terminated entities, information units, informational patterns). The overall hypothesis of LACT requires that the information structure, necessarily signaled by prosody, dominates syntax. The paper gives a qualitative analysis pointing out the strong syntactic reduction, which is claimed to be strictly dependent on its informational organization. Actually the syntax of the text mainly corresponds only to the strict verbal or noun regency occurring within each information unit.
[ "Linguistics & Cognitive NLP", "Syntactic Text Processing", "Linguistic Theories" ]
[ 48, 15, 57 ]
SCOPUS_ID:85135157558
'Assuming our place in the concert of nations': Burundi as imagined in Pierre Nkurunziza's political speeches
Pierre Nkurunziza died in 2020, just a few months short of completing his tenure as the first post-civil war President of Burundi. Critics have cast him as yet another rebel-turned-politician who came to office on a promise of a democratic transformation but became progressively authoritarian, particularly during his third, disputed term in office. As a political figure, however, Nkurunziza remains poorly understood. What kind of a worldview motivated his politics? Drawing on critical discourse analysis, we identify three recurring themes in Nkurunziza's key political speeches: anti-colonialism; unity and self-sufficiency; and discourse around 'politics of a new beginning'. These themes were stable across time, indicating Nkurunziza's consistent worldview, but became more pronounced and radical as he faced growing challenges to his legitimacy from within and without. Far from being confined to rhetoric, the themes also manifested in concrete policy decisions, underscoring the urgent need to take ideology seriously in understanding the political trajectories of African leaders.
[ "Discourse & Pragmatics", "Semantic Text Processing", "Speech & Audio in NLP", "Multimodality" ]
[ 71, 72, 70, 74 ]
SCOPUS_ID:84876034340
'Asylum shopping' in the neoliberal social imaginary
This article critically explores the construction and discursive role of 'asylum shopping' in the cultural politics of asylum in the UK. Despite the unusual combination of a concept predominantly associated with consumerism with one largely associated with human rights or sanctuary, the expression 'asylum shopping' has featured in the mainstream news media and political discourse surrounding asylum and refugee issues since the early 1990s. Drawing upon cultural studies theory, post-Marxist discourse theory and critical discourse analysis, the article argues that the naturalisation of this term has been conditioned by the operation of powerful logics underpinning fundamental insecurities in the identity of the national and neoliberal subject - logics associated with Britain's postcoloniality on the one hand and its neoliberal modernity on the other. While the erosion of collective models of solidarity in favour of entrepreneurialism of the self have provided conditions of possibility for an overwhelmingly negative asylum discourse, outrage at asylum seekers' perceived agency and choice of destination encoded in the notion of 'asylum shopping' have been indexed to nostalgic longings for a more secure national or social identity, as well as deep-seated fears and uncertainties about future prospects in the neoliberal subject. © The Author(s) 2013.
[ "Discourse & Pragmatics", "Semantic Text Processing", "Linguistics & Cognitive NLP", "Linguistic Theories" ]
[ 71, 72, 48, 57 ]
SCOPUS_ID:85059894639
'Aye' or 'No'? Speech-level sentiment analysis of hansard UK parliamentary debate transcripts
Transcripts of UK parliamentary debates provide access to the opinions of politicians towards many important topics, but due to the large quantity of textual data and the specialised language used, they are not straightforward for human readers to process. We apply opinion mining methods to these transcripts to classify the sentiment polarity of speakers as being either positive or negative towards the motions proposed in the debates. We compare classification performance on a novel corpus using both manually annotated sentiment labels and labels derived from the speakers' votes ('aye' or 'no'). We introduce a two-step classification model, and evaluate the performance of both one- and two-step models, as well as the use of a range of textual and contextual features. Results suggest that textual features are more indicative of manually annotated class labels. Conversely, in addition to boosting performance, contextual metadata features are particularly indicative of vote labels. Use of the two-step debate model results in performance gains and appears to capture some of the complexity of the debate format. Optimum performance on this data is achieved using all features to train a multi-layer neural network, indicating that such models may be most able to exploit the relationships between textual and contextual cues in parliamentary debate speeches.
[ "Information Extraction & Text Mining", "Text Classification", "Speech & Audio in NLP", "Sentiment Analysis", "Information Retrieval", "Multimodality" ]
[ 3, 36, 70, 78, 24, 74 ]
SCOPUS_ID:84890771776
'Back to the Rough Ground!' Wittgensteinian Reflections on Rationality and Reason
Wittgenstein does not talk much explicitly about reason as a general concept, but this paper aims to sketch some thoughts which might fit his later outlook and which are suggested by his approach to language. The need for some notions in the area of 'reason' and 'rationality' are rooted in our ability to engage in discursive and persuasive linguistic exchanges. But because such exchanges can (as Wittgenstein emphasises) be so various, we should expect the notions to come in many versions, shaped by history and culture. Awareness of this variety, and of the distinctive elements of our own Western European history, may provide some defence against the temptation of conceptions, such as that of 'perfect rationality', which operate in unhelpfully simplified and idealised terms.© 2008 The Authors.
[ "Reasoning" ]
[ 8 ]
SCOPUS_ID:84859563625
'Before I didn't understand anything about white people, but now, I speak English': Negotiating globally mediated discourses of race, language, and nation
This article explores the ways in which Mexican transmigrants in the USA discursively construct national identities in relation to the mediated message of a television advertisement for an English-language self-study program marketed to Spanish speakers, called Ingles Sin Barreras. Using narrative analysis of the advertisement and critical discourse analysis of a focus group, the author considers the ways in which identities and ideologies of language and language learning come to be intertwined and circulate across national boundaries. © 2012 Copyright Taylor and Francis Group, LLC.
[ "Discourse & Pragmatics", "Semantic Text Processing" ]
[ 71, 72 ]
SCOPUS_ID:85083771985
'Begin the morning with a prayer, and spend the day in temporal affairs': Buddhist ethics in Usun Debeskertu Khan's Homilies
Introduction. The combination of two rules - temporal and religious ones - is typical for Classical Mongolian 'twofold' didactic compositions. Buddhist ethics is a specific moral system represented by a corpus of reputable texts compiled by monastic elites in accordance with their beliefs. Even if some of the works do not focus on religious precepts as such, certain ultimate objectives of the former are explained from the perspective of the Buddhist teaching. And one such didactic composition is The Story of Usun Debeskertu Khan. Its Oirat translation bears different titles, including ones mentioning those are precepts of the Tibetan King Tri Ralpachen (Oir. Usun Debeskertu Khan). The text is compiled from versified homilies that are essentially close enough to oral aphoristic poetry, though distinguished by an explicit form. It covers all spheres of human life and structurally oppose virtuous deeds to sinful ones. The homilies primarily sought to advocate behavioral patterns that would meet socially accepted religious norms. As for instructions dealing with religious practice, those are few enough. However, the bulk of formal temporal guidelines - one way or another - contain religious aspects. Goals and Objectives. The paper aims to introduce into scientific discourse the Oirat text of The Story of Usun Debeskertu Khan and analyze its didactic contents. It is also essential to determine the religious impact and reveal the Buddhist elements. Methods. The article basically employs the historic-philosophical and comparative methods. Research novelty. Despite the wide distribution of the Buddhist ethical text in manuscript copies, its contents have never been analyzed scholarly. Results. The study shows that Buddhist ethics as a system of moral norms is manifested in different texts elaborated by monastic elites pursuant to the Buddhadharma. The composition examined emphasizes no religious aspects, though seeking to guide its readers towards the Buddhist path of salvation. Conclusions. The analysis conducted reveals the religious essence of the homilies, Buddhist ethics constituting the actual background of the formal narrative.
[ "Ethical NLP", "Responsible & Trustworthy NLP" ]
[ 17, 4 ]
SCOPUS_ID:65649153699
'Being Canadian' and 'Being Indian': Subject positions and discourses used in South Asian-Canadian women's talk about Ethnic identity
Ethnic identity descriptions can be viewed as 'subject positions' (Davies and Harr-©, 1990) that are dynamically adopted and discarded for pragmatic purposes through the medium of socialinteraction.Inthe present paper, we use positioning theory to explore the multiple ways our participantsĝ€ "South Asian-Canadian womenĝ€"positioned themselves and others in conversations about their ethnic identity. A discourse analysis of participants' talk revealed a tendency to privilege a 'hybrid' Canadian/South Asian identity over a unicultural one. Moreover, in the rare instances when participants positioned themselves with a unicultural identity, subtle social pressure from conversational partners seemed to induce them to reposition themselves (or others) with a hybrid identity. We conclude by giving possible reasons for such a preference and by discussing the ways in which the current study corroborates and expands on the extant literature. © 2009 SAGE Publications.
[ "Discourse & Pragmatics", "Semantic Text Processing" ]
[ 71, 72 ]
SCOPUS_ID:83155176378
'Bildung' in German human sciences: The discursive transformation of a concept
This article analyses the transformation of the notion of Bildung that is constructed in the German human sciences. From a perspective of field theory and discourse analysis, the article reveals how the notion evolves and stabilizes during a first stage (1810-60), how it comes under pressure because of the contextual changes in a second stage (1860-1960) and how the tension increases before it is resolved by a fundamental change of the traditional notion of Bildung in a third stage (1960-99). © The Author(s) 2011.
[ "Discourse & Pragmatics", "Semantic Text Processing" ]
[ 71, 72 ]
SCOPUS_ID:28244465169
'Blacks and bubbas': Stereotypes, ideology, and categorization processes in restaurant servers' discourse
Individuals employ general, cognitively grounded categorization processes to form expectations for interactions with members of other social groups. Such categorizations sometimes surface in the form of racial, ethnic, or other stereotypes. But although much literature describes and/or tests the cognitive nature of stereotyping and categorization, less investigates how stereotypes and categories are formed in casual interaction, through casual discourse. This article analyzes data from 15 in-depth, semi-structured interviews with restaurant servers to investigate how they categorize customers by drawing on racial stereotypes and stereotypes related to class and/or cultural capital to produce two types of discriminatory discourse: 'racetalk' and what we term 'regiontalk'. Our analyses suggest potential differences in the servers' processes of categorization according to patron type, which we interpret with regard to the larger context of racism and classism in contemporary U.S. society. Copyright © 2005 SAGE Publications.
[ "Semantic Text Processing", "Information Retrieval", "Information Extraction & Text Mining", "Discourse & Pragmatics", "Ethical NLP", "Text Classification", "Responsible & Trustworthy NLP" ]
[ 72, 24, 3, 71, 17, 36, 4 ]
SCOPUS_ID:85148417235
'But what are they?': Zine-making and invitational creative practice in an undergraduate creative writing class inspired by the work of Lynda Barry
This illustrated article offers a record of work done for a third-year undergraduate module called Creative Writing and the Self, as part of the Creative and Professional Writing programme at the University of the West of England, Bristol, in the autumn term of 2021, where students created zines as a record of their student experiences, which corresponded to the years of the pandemic. The article considers the students' creative process and what is communicated in the zines, pages from which illustrate the article, informed by the methodology of Lynda Barry. The module offered the opportunity for staff and students to use a more open-ended rhetoric than elsewhere on their course(s), and the article considers this alongside the pedagogical implications of a hybrid and multimodal form and approach, of telling stories using text and images, inviting acknowledgement and observation, rather than judgement.
[ "Visual Data in NLP", "Multimodality" ]
[ 20, 74 ]
SCOPUS_ID:33644942483
'But where are our moral heroes?' An analysis of South African press reporting on children affected by HIV/AIDS
Messages conveyed both explicitly and implicitly in the media play an important role in shaping the public's understanding of issues, as well as in shaping associated policy, programmes and popular responses to these issues. This paper applies discourse analysis to a series of articles about children affected by HIV/AIDS published in 2002/2003 in the English-language South African press. The analysis reveals layers of moral messaging present in the reporting, the cumulative effect of which is the communication of a series of moral judgements about who is and who is not performing appropriate roles in relation to children. Discourses of moral transgression, specifically on the part of African parents and 'families'for failing in their moral responsibilities towards their children, coalesce with discourses on anticipated moral decay among (previously innocent) children who lack their due care. A need for moral regeneration among South Africans (but implicitly black South Africans) contrasts with accolades for (usually white), middle-class individuals, who, it is implied, have gone beyond their moral duty to respond. The article argues that in each instance the particular moralism is questionable in light of both empirical evidence and the principles of human dignity that underlie the South African constitution. Children - and particularly 'AIDS orphans' - are often presented in the press as either quintessential, innocent victims of the epidemic or as potential delinquents. While journalists' intentions are likely to be positive when representing children in these ways, the paper argues that this approach is employed at a cost, both to the public's knowledge and attitudes around the impact of HIV/AIDS, and, more importantly, to the lives of children affected by the epidemic. Copyright © NISC Pty Ltd.
[ "Discourse & Pragmatics", "Semantic Text Processing" ]
[ 71, 72 ]
SCOPUS_ID:84859065760
'But why do I have to take this class?' The mandatory drama-ESL class and multiliteracies pedagogy
This article seeks to understand the role of drama pedagogy in second/additional language learning with data drawn from a school-based ethnographic study of English language learners taking a drama-English as a Second Language (ESL) course. Being aware that all drama teaching does not automatically lead into improvement in language learning, it carefully explores the experiences of a group of English language learners taking a mandatory drama-ESL class after having passed a proficiency exam. Drawing from her data and from theoretical work in drama education, second language education and postcolonial discourse, the author proceeds to examine two aspects of multiliteracies pedagogy: situated practice and multimodality. One key finding is that, despite initial resistance from the majority of the English language learners about taking this mandatory class, the drama pedagogy used in this classroom drew on students' personal and cultural experiences in the creation of identity texts and thereby provided room for a situated practice as well as multimodal representations of meaning. This process of creating performance-based identity texts, the author argues, cognitively engaged students, provided room for identity investment and, therefore, despite initial challenges, helped many students with their linguistic and social performances. © 2011 Copyright Taylor and Francis Group, LLC.
[ "Multimodality" ]
[ 74 ]
SCOPUS_ID:85053646750
'Call for purge on the people traffickers': An investigation into British newspapers' representation of transnational human trafficking, 2000-2016
Gregoriou and Ras draw on corpus linguistics and critical discourse analysis to examine a 61.5 million-word corpus of articles published by UK newspapers between 2000 and 2016, and on qualitative critical discourse analysis of a sixty-seven-article sample corpus in depth. Both approaches analyse the naming and describing of victims and traffickers, metaphors, transitivity, and speech and writing presentation, while the in-depth qualitative approach furthermore analyses the text (images) (multi)modally. Their findings conclude that trafficking for sexual exploitation is over-reported compared to other forms of trafficking, and that victims are generally presented as young, female, and vulnerable. As a result, non-stereotypical victims, of crimes like forced begging and domestic servitude, are not readily recognised as victims, and thereby are deprived of opportunities for assistance.
[ "Discourse & Pragmatics", "Semantic Text Processing", "Representation Learning" ]
[ 71, 72, 12 ]
SCOPUS_ID:84863833874
'Cantonese is not a dialect': Chinese netizens' defence of Cantonese as a regional lingua franca
This article reports on an inquiry into Chinese netizens' online discussions related to the 'Protecting Cantonese Movement' in Guangzhou, Guangdong Province, on the Chinese mainland. It interprets the ideological discourses used by Chinese netizens in online discussions to protect the status of Cantonese, a regional variety of the Chinese language. These netizens were found to have drawn on the international prestige and traditional heritage of Cantonese in arguing for maintaining its status as a regional lingua franca. Drawing on research on the individualisation of society in China, this article contends that these netizens may be seen to be recontextualising the political establishment's discourses and appropriating them as powerful weapons in defence of their linguistic rights. It was also found in the inquiry that non-Cantonese-speaking migrants were problematised by the netizens as a cause of the predicament of Cantonese, creating a significant challenge for policymakers and language educators in their efforts to create a 'harmonious' society on the Chinese mainland. One may argue that harmony can be achieved through respecting individuals' linguistic rights. © 2012 Copyright Taylor and Francis Group, LLC.
[ "Discourse & Pragmatics", "Semantic Text Processing" ]
[ 71, 72 ]
SCOPUS_ID:84923246668
'Case Relations' in Lao, A Radically Isolating Language
This article examines data from Lao, a radically isolating Southwestern Tai language spoken in Laos, Thailand, and Cambodia, and asks how speakers of such a language might cope without case. Lao is like Mandarin, Thai, Vietnamese, and Riau Indonesian in exemplifying the extreme of pragmatically oriented grammar. Where case marking simply distinguishes who from whom, it is mostly dispensable, thanks to the richness of pragmatics. Moreover, for more 'expressive' functions of case marking, where features of transitivity are manipulated for expressive or information-structural effect, Lao finds constructional means to treat certain arguments in special ways, thereby explicitly marking non-redundant semantic information in case-like ways. This article also examines patterns of argument-predicate relations, focusing on monovalent predicates, symmetric and other non-oriented bivalent predicates, and asymmetric bivalent predicates. Finally, it considers the expressive functions which case marking might perform, that is where special treatment of one or another argument serves to manipulate semantic distinctions in the construal of event structure and participant involvement.
[ "Discourse & Pragmatics", "Semantic Text Processing" ]
[ 71, 72 ]
SCOPUS_ID:84879504334
'Categories We Share': Mobilising Common In-groups in Discourse on Contemporary Immigration in Greece
Prejudice reduction has been an important concern within social psychology both in theory and applied research. According to the premises of Social Identity Theory, redrawing of the category boundaries is often considered a necessary step in order to battle prejudice, because in-group favouritism when the category boundaries change is diffused to the previously distinct identities. The present paper offers a review of the relevant research, and following a discourse analytic perspective argues that recategorisation can also be viewed as a rhetorical resource that people use in verbal interaction in order to achieve certain rhetorical ends. This point is exemplified using interview data from Greece with Greek participants who mobilise common in-groups between themselves and the immigrants in Greece. Different common in-groups were mobilised on the basis of common human nature, common ethnic descent and through the use of the common experience of migration that many Greek people have because Greece has been an emigrant sending country for the biggest part of the 20th century. Occasionally, these category constructions were used to differentiate between immigrants of different ethnic descent claiming that only certain immigrant groups can integrate to Greek society, whereas on other instances, these common in-groups were used in order to inoculate speakers of accusations of prejudice. © 2012 John Wiley & Sons, Ltd.
[ "Discourse & Pragmatics", "Semantic Text Processing", "Linguistics & Cognitive NLP", "Linguistic Theories" ]
[ 71, 72, 48, 57 ]
SCOPUS_ID:84859494306
'Challenge' or 'collaboration' social interaction and recontextualization: McDonald's CSR report
This paper is concerned with analyzing McDonald's language use in its response to two different social issues: one regarding employees ('McJob') and the other regarding the environment (culpability for deforestation). It discusses and compares these two different social interactions in terms of discourse formations (DFs) - [MCJOB] and [DEFORESTATION] - based on McAndrew's [2001. Intertextuality, heteroglossia and systemic functional linguistics: A framework for analyzing ideology (Ph.D. thesis, Macquarie University, NCELTR, Sydney); 2004. Towards a framework for analyzing ideology: Applying intertextuality, heteroglossia and systemic functional linguistics. In M. Putz, J. Aerselaer & T. van Dijk (Eds.), Discourse, and social practice. Frankfruit: Peter Lang] framework for analysis of ideology. In particular, the meanings constructed in each of the DFs are realized by the analysis of social actions and social actors using Van Leeuwen's [1993. Language and representation - The recontextualization of participants, activities and reactions (PhD thesis, University of Sydney); 1995. Representing social action. Discourse & Society, 6(1), 81-106; 1996. The representation of social actors. In C.R. Caldas-Coulthard & M. Coulthard (Eds.), Texts and practices: Readings in critical discourse analysis. London & NY: Routledge] representational model. It was found that McDonald's responded differently to the criticism in the two different areas. McDonald's challenges the term 'McJob' by (re)constructing meanings to disagree with the term, but it collaborates with Greenpeace, which had published two reports blaming McDonald's contribution to deforestation in the Amazon area. However, simultaneously, through intertextual meaning-making, McDonald's commonly attempts to recontextualize its social practices by promoting its socially responsible business practices. © 2012 Copyright Taylor and Francis Group, LLC.
[ "Discourse & Pragmatics", "Semantic Text Processing" ]
[ 71, 72 ]
SCOPUS_ID:33748756790
'Children are just lingual': The development of phonology in British Sign Language (BSL)
This paper explores three universal tendencies in spoken language acquisition: consonant and vowel harmony, cluster reduction and systemic simplification, using a corpus of 1018 signs from a single child exposed to British Sign Language (BSL) from birth. Child signs were recorded from naturalistic deaf parent-deaf child interaction between the ages of 19-24 months. Child errors were analysed by handshape, movement and location segments, as well as the accurate production of prosodic features, using an autosegmental phonology approach. Unadult like forms at this age were observed with 41% of handshapes, 45% of movements and 25% of locations. There were 47% of signs produced with unadult like prosodic features. Analysis of the results concludes that early child signing broadly follows proposed universal tendencies in language acquisition. © 2006 Elsevier B.V. All rights reserved.
[ "Phonology", "Syntactic Text Processing" ]
[ 6, 15 ]
SCOPUS_ID:84988849482
'Clustering' documents automatically to support scoping reviews of research: a case study
BACKGROUND: Scoping reviews of research help determine the feasibility and the resource requirements of conducting a systematic review, and the potential to generate a description of the literature quickly is attractive.
[ "Information Extraction & Text Mining", "Text Clustering" ]
[ 3, 29 ]
SCOPUS_ID:84897519916
'Code switching' and linguistic theory: The Italian-Romance situation
This paper looks at the main aspects involving relationships between code switching phenomena and linguistic theory in the Italian language-dialect contact situation. The analytical remarks are based on two different perspectives: the pragmatic-functional one, related to sociolinguistic theories (§ 2.), and the grammatical one (§ 3.). After pointing out the notion of dialect in the Italo-Romance context (§ 1.), the first section deals with applicability of pragmatic-functional models, such as the Interpretive Sociolinguistics one and the Markedness Model by Carol Myers-Scotton, to code switching between Italian and dialect. We discuss social and communicative meanings of code switching, investigating its function as identity carrier and its relationship with linguistic repertoires typology and dialect decay processes. The second section focuses on intra-sentential code switching grammatical constraints; in particular we present and discuss the notion of matrix language and the two main principles governing Myers Scotton's Matrix Language Frame Model (Morpheme Order Principle and System Morpheme Principle). The concept of hybridism and differences between hybridism and borrowing are also discussed at length. In the last section (§ 4.) we underline some methodological similarities which could lead to an integrated approach to code switching phenomena.
[ "Code-Switching", "Multilinguality", "Linguistics & Cognitive NLP", "Linguistic Theories" ]
[ 7, 0, 48, 57 ]
SCOPUS_ID:84902256738
'Coherence at Last': Percival Gurrey's Contribution to English
Reacting to incoherent English teaching in the 1930s, Percival Gurrey probed the psychological processes involved in literary appreciation. He sought ways of teaching poetry that avoided lifeless tasks such as labelling 'poetic devices'. Later, in the 1950s, he wrote about the processes involved in learning to write. At a time when psychology dominated educational discourse, he was in touch with new developments in both literary criticism and linguistics, and he evolved a bridging theory of language, learning and development that spanned philosophy, psychology and literary theory. I connect Gurrey's sense of wholeness of response to his reading in Coleridge. By returning to debates surrounding Coleridge I show how a new appreciation of what language accomplishes emerged. Gurrey discovered that writing has a central place in children's development as a whole. From such hard-won discoveries a unified theory of development in talking, reading and writing duly emerged. © 2014 © 2014 The editors of Changing English.
[ "Linguistics & Cognitive NLP", "Linguistic Theories" ]
[ 48, 57 ]
SCOPUS_ID:76949096731
'Coming out' on the spectrum: Autism, identity and disclosure
Much has been written by queer theorists about the personal and political ramifications of being out of the closet, and connections with experiences of disclosure for those with 'hidden' health conditions have been made by researchers studying critical geographies of disabilities and chronic illness. To date, however, the impact of such issues for those on the autism spectrum (AS) has received comparatively little attention. Popular re-presentations of AS suggest disclosure is irrelevant for those assumed so obviously different and unlikely to pass as 'normal.' However, AS authors reveal a broad spectrum of experience indicating that concealment and disclosure are complex and selective strategies of information and identity management. Applying discourse analysis to AS autobiographies and personal narratives, this paper explores four sense-making discourse clusters, or repertoires, that emerge from the texts under study: a 'keeping safe' repertoire, which addresses protective strategies in disclosure and coming out; a 'qualified deception' repertoire, which relates to the complexities of non-disclosure; a 'like/as resistance' repertoire, which captures the tendency of AS authors to position their individual and collective experiences of coming out on the spectrum as analogous to the process of coming out for other marginalized groups, most notably gay and Deaf communities; and an 'education' repertoire, which contributes to the project of building a community to come out to. Each of these repertoires is situated within the broader literature in social and cultural geography and critical disability studies. © 2010 Taylor & Francis.
[ "Discourse & Pragmatics", "Semantic Text Processing" ]
[ 71, 72 ]
SCOPUS_ID:84873156877
'Committed to the ideals of 1916'. The language of paramilitary groups: The case of the Irish Republican Army
The objective of this article is to describe and understand the language of paramilitary groups in the Northern Irish context, taking statements issued by the Irish Republican Army (IRA) as an example. In order to do so, we depart from a broad understanding of political discourse. So as to understand how beliefs, actions and the IRA existence are legitimised in those statements, Text-World Theory is combined with critical discourse analysis approaches. Chilton's notion of 'discourse worlds' is considered the main legitimising strategy, and these - understood as mental representations - are identified by relying on a micro-analysis of linguistic choices. Thirty-nine IRA statements delivered after the 1998 Agreement have been analysed in order to understand both synchronic responses to given socio-political events and diachronic changes related to the advances of the Northern Irish Peace process. © 2013 Copyright Taylor and Francis Group, LLC.
[ "Discourse & Pragmatics", "Semantic Text Processing" ]
[ 71, 72 ]
SCOPUS_ID:85084284818
'Computing with Words'- based Semantic Similarity Measure for Adjective Phrases
This paper addresses the semantic similarity computation between adjective phrases like 'more or less tall and not very short', 'somewhat tall and not extremely tall', 'very short and not very very tall' etc. The proposal finds application in scenarios where human perception based inputs are involved, such as in processing product reviews, feedbacks, survey responses, social media posts etc. To be able to adequately access the semantic spectrum, we employ the Computing with words (CWW) based technique in formulating the similarity measure. Particularly, we use the concepts of a linguistic variable and linguistic hedge. The implementation results report similarity scores that are in agreement with human reasoning.
[ "Semantic Text Processing", "Semantic Similarity" ]
[ 72, 53 ]
SCOPUS_ID:85140756969
'Cool glasses, where did you get them?' Generating Visually Grounded Conversation Starters for Human-Robot Dialogue
Visually situated language interaction is an important challenge in multi-modal Human-Robot Interaction (URI). In this context we present a data-driven method to generate situated conversation starters based on visual context. We take visual data about the interactants and generate appropriate greetings for conversational agents in the context of HRI. For this, we constructed a novel open-source data set consisting of 4000 URI-oriented images of people facing the camera, each augmented by three conversation-starting questions. We compared a baseline retrieval-based model and a generative model. Human evaluation of the models using crowdsourcing shows that the generative model scores best, specifically at correctly referencing visual features. We also investigated how automated metrics can be used as a proxy for human evaluation and found that common automated metrics are a poor substitute for human judgement. Finally, we provide a proof-of-concept demonstrator through an interaction with a Furhat social robot.
[ "Natural Language Interfaces", "Visual Data in NLP", "Multimodality", "Dialogue Systems & Conversational Agents" ]
[ 11, 20, 74, 38 ]
SCOPUS_ID:84856879505
'Cos girls aren't supposed to eat like pigs are they?' Young women negotiating gendered discursive constructions of food and eating
While psycho-medical understandings of 'eating disorders' draw distinctions between those who 'have'/'do not have' eating disorders, feminist poststructuralist researchers argue that these detract from political/socio- cultural conditions that invoke problematic eating and embodied subjectivities. Using poststructuralist discourse analysis, we examine young women's talk around food and eating, in particular, the negotiation of tensions arising from derogating aspects of hetero-normative femininities, while accounting for own 'feminine' practices (e.g. 'dieting') and subjectivities. Analysis suggested that eating/dieting was accounted for by drawing upon neo-liberalist discourses around individual choice; however, these may obscure gendered, classed and racialized power relations operating in local and wider contexts. © The Author(s) 2011.
[ "Discourse & Pragmatics", "Semantic Text Processing" ]
[ 71, 72 ]
SCOPUS_ID:85114873342
'Could I have an appointment for a viewing?' Language-based discrimination and apartment searches with different accents in Germany
This paper represents a novel approach to the study of discrimination in the housing market. Beginning with detailed discourse analyses of several excerpts of apartment application conversations, it highlights how Standard German, Standard American and Turkish accents interact and how power relations are reproduced on a micro-level through interruptions and repair initiations. Furthermore, it includes the statistical report of the viewing appointments resulting from almost 300 phone calls placed in four different city districts of the city of Bremen, Germany, with Turkish, Standard American, and German names and accents. The neighborhoods, not the city per se, are a crucial aspect for predicting linguistic discrimination: in the more prestigious neighborhood, Turkish accented callers had significantly lower chances of receiving a viewing. In all but one city district, the Standard German callers received the most viewing appointments, and the American English accented callers had more chances than the Turkish callers speaking Standard German.
[ "Discourse & Pragmatics", "Semantic Text Processing", "Information Retrieval" ]
[ 71, 72, 24 ]
SCOPUS_ID:85119204312
'Could You Describe the Reason for the Transfer?': A Reinforcement Learning Based Voice-Enabled Bot Protecting Customers from Financial Frauds
With the booming of the Internet finance and e-payment business, telecom and online fraud has become a serious problem which grows rapidly. In China, 351 billion RMB (approximately 0.3% of China's GDP) was lost in 2018 due to telecommunication and online fraud, influencing tens of millions of individual customers. Anti-fraud algorithms have been widely adopted by major Internet finance companies to detect and block transactions induced by scam. However, due to limited contextual information, most systems would probably mistakenly block the normal transactions, leading to poor user experience. On the other hand, if the transactions induced by scam are detected yet not fully explained to the users, the users will continue to pay, suffering from direct financial losses. To address these problems, we design a voice-enabled bot that interacts with the customers who are involved with potential telecommunication and online frauds decided by the back-end system. The bot seeks additional information from the customers through natural conversations to confirm whether the customers are scammed and identify the actual fraud types. The details about the frauds are then provided to convince the customers that they are on the edge of being scammed. Our bot adopts offline reinforcement learning (RL) to learn dialogue policies from real-world human-human chat logs. During the conversations, our bot also identifies fraud types every turn based on the dialogue state. The bot proposed outperforms baseline dialogue strategies by 2.8% in terms of task success rate, and 5% in terms of dialogue accuracy in offline evaluations. Furthermore, in the 8 months of real-world deployment, our bot lowers the dissatisfaction rate by 25% and increases the fraud prevention rate by 135% relatively, indicating a significant improvement in user experience as well as anti-fraud effectiveness. More importantly, we help prevent millions of users from being deceived, and avoid trillions of financial losses.
[ "Speech & Audio in NLP", "Natural Language Interfaces", "Dialogue Systems & Conversational Agents", "Reasoning", "Multimodality" ]
[ 70, 11, 38, 8, 74 ]
SCOPUS_ID:85030543746
'Culture' and 'the Cultural': On the reflexivity of a popular concept
This paper discusses the interpretive complexity of 'culture', which is at the same time the central theoretical concept of 'cultural linguistics' (kulturwissenschaftliche Linguistik) as well as the object of its investigation, i.e., a conceptual term and a discursive phenomenon. This results in ambiguous conceptualizations: In culture-theoretical work, 'culture' is usually described as being 'dynamic', 'fluid', 'procedural' and 'context-dependent', but in discourse, culture is often, if not predominantly, perceived as something 'static', 'fixed', 'essential' and 'general'. From this ambiguity, many methodological questions emerge: Is it advisable to rely on a concept that has such an enormous ideological load (and that, moreover, is increasingly used in recent neo-nationalist and even neo-racist propaganda)? Does culture analysis willy-nilly perpetuate problematic essentialist framings while relying on 'culture' as a descriptive concept? Is it methodologically sensible to use concepts for description and explanation of discourses that are actually products of the discourses that are supposed to be explained (cf. Foucault 2001 [1968])? In other words: is 'cultural linguistics' a circular venture if it explains cultural discourses on the background of the concept of culture? Is the discrepancy between dynamic and static, 'descriptive' and 'prescriptive' concepts of culture to be solved, and how? Should we try to de-ideologize 'culture', as some culturally oriented scholars have proposed? Or should we rather drop the concept, as other scholars (particularly from a post-colonial point of view) have suggested? The paper discusses these questions, argues for more sensitivity vis-à-vis the ideological contextualization of 'culture' and suggests differentiating ideological fixations of 'culture' from the discursive processes and negotiations that constitute 'the cultural'.
[ "Discourse & Pragmatics", "Explainability & Interpretability in NLP", "Semantic Text Processing", "Responsible & Trustworthy NLP" ]
[ 71, 81, 72, 4 ]
SCOPUS_ID:77949499565
'Curvy, hunky, kinky': Using corpora as tools for critical analysis
This article, an expanded version of an oral presentation in 1999, uses corpus methodology as a research tool to investigate how social actors are classified in the public discourse of the media, with lexis as our point of entry. Our main focus is the nature of the labels which provide categorization, especially of gender relations. Our main claim is that uses of premodification associated with the two types of newspapers in Britain and their lexical choices produce differential judgmental stances that have social effects. In the first of two complementary studies, we discuss the adjective lexicon of the tabloid press in comparison with quality newspapers, with curvy, hunky and kinky as exemplars with respect to sexualization and the construction of gender. In our second study, we discuss adjectival premodification of man, woman, girl and boy in tabloids and broadsheets: Our findings show that the media categorizes people through very specific points of view and values not always apparent to a non-critical reader. Collocational patterns undoubtedly reveal societal and sociolectal attitudes, especially in terms of stereotypes of gender, sexualization, age and behaviour. Our main aim, therefore, is to show that corpus studies can help to deconstruct hidden meanings and the asymmetrical ways people are represented in the press. © The Author(s) 2010.
[ "Information Retrieval", "Text Classification", "Information Extraction & Text Mining" ]
[ 24, 36, 3 ]
SCOPUS_ID:85106386604
'Devotio' Lindsay: Devotional practices and visual culture on the internet in the 21<sup>st</sup> century
Studies about the theory of religious image and visual pragmatics consider that nowadays the practice of “image-assisted devotion” has fallen into disuse and, therefore, images for affective purposes between human beings and deities are no longer in use. As the Internet is considered the new space where cultural dynamics are created, this article traces within the Lindsay Lohan's Last Fan web performance (2012-2014) if these visual needs have been extinguished. Using as an analytical tool the procedure of image-assisted meditation: -contemplation, edification, empathy, and imitation- similarities are discovered regarding the devotional process. The search for contact with ineffable entities persists but with different images, referents, genres, and ideologies.
[ "Discourse & Pragmatics", "Visual Data in NLP", "Semantic Text Processing", "Multimodality" ]
[ 71, 20, 72, 74 ]
SCOPUS_ID:62349120743
'Diversity' as enacted in US immigration politics and law: A corpus-based approach
As a national institution, the law is held to be above prejudice and bias (Gibbons, 1994). However, in recent years, as public awareness of 'homeland security' has been heightened and borders have been tightened, the laws regarding immigration have come under close scrutiny. This article explores the language of immigration laws that target the notion of 'diversity' and the political contexts in which those laws were created. More specifically, it addresses the following questions: Is there prejudice or bias in US immigration laws? If so, from where do those negative ideologies arise? And, finally, what does 'diversity' actually mean in the context of US immigration? This analysis, which utilizes the tools of corpus and appraisal analysis, shows how contradictory positions surrounding the ideology of diversity exist within seemingly positive political discourses and demonstrates how those underlying judgements about diversity interact with and shape the very language of the laws that ensue. Copyright © 2009 SAGE Publications.
[ "Discourse & Pragmatics", "Semantic Text Processing" ]
[ 71, 72 ]
SCOPUS_ID:79960438941
'Doing family therapy': A Foucauldian discourse analysis
This paper draws on a study that asked twenty family users about their first session of family therapy. Analyses of the interviews indicated that families entered therapy with a pre-existent knowledge about therapy, which did not always chime with those of professionals and which positioned speakers in ways which governed their expectations and perceptions of therapy. This paper, therefore, is concerned with the acquisition and deployment of knowledge: specifically, the knowledge involved in being a user of family therapy. Three key discourses were identified through this analysis: medical, counselling and consumerist. We aim to illustrate how these discourses served as a resource for members of the family in constructing therapists, therapy and themselves in relation to their experience. The examination of the rhetorical, ideological and practical effects of the positions chosen and the objects constructed, in terms of how speakers wanted to present themselves, with what enhanced or diminished status as patients, shows users actively engaging with the power of therapeutic institutions. Users' accounts suggest that while most speakers felt anxious about the prospect of therapy, there were clear differences in overall satisfaction/dissatisfaction with the experience according to the synchrony between speakers' construction of the therapist, and themselves as client/ patients. Those who seemed to take a traditional view of therapy within a medical discourse valued therapists who offered diagnosis and a cure; those who sought and experienced a counselling relationship with the therapist found their experience to be constructive in terms of enhanced self-knowledge. We believe that the research findings discussed in this paper have implications for family therapists in accommodating to parental and child positions to maximise the effectiveness of therapy and so minimise drop-out. © 2010 Taylor & Francis.
[ "Discourse & Pragmatics", "Semantic Text Processing" ]
[ 71, 72 ]
SCOPUS_ID:34548221456
'Don't get caught out': Pragmatic and discourse features of informational and promotional texts in international healthcare insurance
This paper examines the pragmatic and discourse features of a number of multimodal international healthcare insurance texts. The texts contain specialized language related to the provision and treatment of healthcare and to insurance coverage and are aimed at a fairly well-defined target readership, many of them actual or potential expatriates. They thus serve a useful function, that of providing the readership with an informed and comprehensible guide to the services and products available. Some of them have a more strictly promotional aspect, that of advertising specific insurance policies. In particular, the paper focuses on the language choices favored by these texts and the strategies they serve, noting that the distinction between their informative and promotional dimensions is not always clear-cut. It further suggests that the writers of these texts make a specific use of discourse features and languaging strategies in order to achieve their rhetorical purposes. © Walter de Gruyter 2007.
[ "Multimodality" ]
[ 74 ]
SCOPUS_ID:84954370774
'Don't give us an assignment where we have to use spinach!': Food choice and discourse in home and consumer studies
The aim of the study was to describe classroom Discourses about vegetables during the planning, cooking, eating and evaluation of meals in the Swedish school subject Home and Consumer Studies. Fifty-nine students and five teachers were recruited from five northern Swedish villages and towns, and then observed, recorded and in some cases video-taped during lessons that took place between 2010 and 2012. Based on 56 instances of talk about vegetables, four Discourses were identified and related to the three aspects of Belasco's culinary triangle of contradictions: identity, responsibility and convenience. The results indicated that the identity-based sensory and cultural Discourses sometimes clashed with the more responsibility-oriented health and evaluation Discourses. The health Discourse was only used when there was an element of evaluation, with assignments connected to grades. In all other cases, the sensory and cultural Discourses guided vegetable use. Sometimes different sensory or cultural assumptions could clash with each other, for example when the teacher insisted on the use of a specific recipe regardless of a student's taste preferences. Since these preferences did not always harmonize with curricular demands for responsibility, there might be a risk of basing grades on aspects of students' identity. Alternatively, students might feel constrained to argue against their own identity in order to be favourably evaluated. Then again, if teachers always bow to student tastes, this limits their chances of learning about food and physical health. Viewing the dilemma through the lens of the culinary triangle of contradictions may help teachers and researchers develop teaching methods that take all aspects of food choice into account.
[ "Discourse & Pragmatics", "Semantic Text Processing", "Responsible & Trustworthy NLP" ]
[ 71, 72, 4 ]
SCOPUS_ID:84909223416
'Ducks' and 'parrots': Elaboration, duplication and duplicity in a cartoon discourse completion test
Interviews, questionnaires and tests may fail as evaluations of pragmatic competence in a foreign language because they cannot reproduce contexts in sufficient detail and/or because they cannot constrain discourse options without contaminating the learner response. In this study a cartoon Discourse Completion Test (DCT) prompt was used, providing a pictorial context which constrains the response but allows the learner freedom to elaborate language. Data gathered from 505 Korean learners of English showed that picture description in the written mode provided the most elaborated language whereas written dialogue was more lexicalised. There was a realistic 'opening up' of discourse options as the cartoon progresses, and a realistic 'narrowing down' of discourse options within each exchange, as the initiating utterance within each frame constrained the rejoinder. But the data also revealed two less realistic ways of coping with a pragmatic task: 'ducking', in which information is ignored or reinterpreted in ways that are easier to deal with, and 'parroting', where language in one frame of the cartoon by one speaker is recycled in another frame by another. Although these blur the effectiveness of the instrument as a summative test, they offer interesting possibilities for formative testing and teaching. © 2002, Taylor & Francis Group, LLC.
[ "Discourse & Pragmatics", "Semantic Text Processing" ]
[ 71, 72 ]
SCOPUS_ID:85033584561
'Eating regret and seeing contempt' - A cognitive linguistic approach to the language of emotions in Igala (Nigeria)
This book deals with the expression of emotions in Igala - a Nigerian minority language with about two million speakers - from a Cognitive Linguistics perspective. The author investigates the network of the Igala language of emotions using the Natural Semantic Metalanguage (NSM) framework as developed by Anna Wierzbicka and her collaborators. The claims of adequacy of NSM in its present form for the analysis of emotion concepts universally will be challenged. This is due to the way emotions are conceptualized in the Igala language and due to the absence of certain semantic primes which are considered necessary for the analysis of emotions.Therefore, this study presents a testing ground for further development of the framework as well as a peek into research in African languages.
[ "Emotion Analysis", "Sentiment Analysis" ]
[ 61, 78 ]
SCOPUS_ID:84875701190
'Effective learning through e-learning system: DEA Approach': HCI technologies for education
The paper investigates the relative efficiency of visual and phonetic features of e-learning system for student's learning effectiveness leading to better academic achievement and employability. The e-learning system is a form of human computer interaction process build and transferable skills through computer. The features used in e-learning system to sustain student's long span of focus attention result in retention of information in long-term memory, which can be utilized later for problem solving and situation. After experimentation of e-learning system along with traditional classroom teaching, the data collected are analyzed through DEA Approach. The findings indicate improvement in academic scoring and employability skills after e-learning system implementation because of on-hand exposure, any time discussion form and open resources. Thus visual and phonetic features applied for study can be utilized by others for better e-learning interface designing. © 2012 IEEE.
[ "Visual Data in NLP", "Phonetics", "Syntactic Text Processing", "Multimodality" ]
[ 20, 64, 15, 74 ]
SCOPUS_ID:85082320548
'Effortful', 'needy' and 'freeloader': Constructions of unemployed people's deservingness in Finnish parliamentary discussions
This article provides a rhetorical discourse analysis of constructions of unemployed people's deservingness. Data consist of transcripts from Finnish parliament members debating the 'Activation Model for Unemployment Security', from December 2017. In the analysis, three discursive constructions of unemployed people's deservingness were identified: an 'effortful citizen lacking control', a 'needy citizen deserving the welfare state's reciprocal acts' and an 'undeserving freeloader in need of an attitude adjustment'. Analysis focuses on how deservingness and undeservingness are rhetorically accomplished and treated as factual in parliament members' accounts. The analysis pays particular attention to the question of how speakers build factuality through the management of categories, extreme case formulations, 'truth talk' and maximisation and minimisation strategies. The results reflect the negotiated nature of deservingness as well as varying constructions of unemployed people's responsibility in the contemporary Nordic welfare state context.
[ "Discourse & Pragmatics", "Semantic Text Processing" ]
[ 71, 72 ]
SCOPUS_ID:78751704835
'Energy security' and 'climate change': Constructing UK energy discursive realities
Recently, in the United Kingdom, two issues have dominated the energy policy agenda: effective climate change mitigation and energy security. Whilst evolving government policy has led to government support for new build nuclear power as part of the nation's future energy mix, limited attention has been devoted to examining how arguments were constructed to lead 'naturally' to nuclear new build as an option for addressing these two issues. Using Critical Discourse Analysis this paper analyses the struggles within the climate change mitigation and energy security discourses in generating and/or replacing meanings. In particular, it examines how construction of the dominant (hegemonic) discourses led 'naturally' to the necessity of new build nuclear power. This paper draws upon 24 stakeholder interviews to examine the hegemonic and counter-hegemonic discourses. It outlines how climate change and energy security were perceived as motivators for energy policy; it shows how the combination of the dominant construction of climate change as an environmental issue and the construction of energy security as a '. gas gap' lent weight to the argument for nuclear new build. Struggling against these discourses is a counter-hegemonic discourse centred around climate change as a symptom of unsustainability and energy security as a lack of energy diversity. The latter, rather than 'naturally' proposing an urgent need for nuclear new build, lead to the argument for readdressing the focus of, and use of resources by, society - reducing energy demand and increasing energy supply diversity. © 2010 Elsevier Ltd.
[ "Discourse & Pragmatics", "Semantic Text Processing" ]
[ 71, 72 ]
SCOPUS_ID:84874319790
'English... it's part of our blood': Ideologies of language and nation in United States Congressional discourse
Utilizing a combination of corpus linguistics and critical discourse analysis, the article describes ideologies of United States (U.S.) national identity based in civic and ethnic nationalisms. These ideologies circulated during debates and hearings in the U.S. Congress concerning the 2006 reauthorization of provisions of the Voting Rights Act requiring that multilingual voting materials be made available for language minorities in certain jurisdictions. The article argues that although those legislators supporting the use of multilingual voting materials construct an ideology of civic nationalism that promotes inclusion of language minorities, they are still greatly influenced by the dominant language ideology. Those opposing multilingual voting materials construct a U.S. national identity that places stringent requirements on newcomers to assimilate and learn English. These legislators simultaneously attempt to avoid the suspicion of ethnic nationalist ideologies by arguing that English is a neutral entity, detaching it from its dominant linguistic community. The findings suggest that despite Congressional support for multilingual voting materials, the language ideologies present in Congress are not necessarily favorable to the rights of language minorities. These findings have language policy implications in that they suggest opposition to a right to language for linguistic minorities in the U.S. © Blackwell Publishing Ltd. 2013.
[ "Discourse & Pragmatics", "Semantic Text Processing", "Multilinguality" ]
[ 71, 72, 0 ]
SCOPUS_ID:85090387316
'Envision': Visual Representation System
The use of technology is increasing day by day in most of the possible areas. Education is one of the area in which technology has helped students in understanding the basic concepts and theories related to their studies. It has been observed that the primary students face problem in understanding some of basic concepts like arithmetic word problems. Envision is a system that incorporates natural language text input to construct scenes in real-time. Such type of system forges the way for systems in which interaction with and construction of graphical media can be as natural as communicating via text. System aims at taking the basic arithmetic word problems of primary section as input and depicting a solution as an image on the output screen. Natural Language Processing (NLP) is used to process the text given by user and to find the objects to create the scene. The image dataset used in this process will grow as the new objects are encountered in the word problem. The system produces photo-realistic images based on the input text using Generative Adversarial Networks (GAN) and helps students to realize what exactly the sentence means. Low resolution images are formed using Stage-I GAN by sketching the primal shape and colours of the object based on the given text description. The Stage-II GAN processes Stage-I outcome and generates high resolution images with photo-realistic information. The system will also be useful for teachers and parents to teach their students/children in an effective way.
[ "Visual Data in NLP", "Semantic Text Processing", "Robustness in NLP", "Representation Learning", "Responsible & Trustworthy NLP", "Multimodality" ]
[ 20, 72, 58, 12, 4, 74 ]
SCOPUS_ID:85082812154
'Eucalypt-free-zone' municipalities: Analysis of the local political actors in Galicia, Spain
The controversy around the eucalypt plantations transcends in Galicia its forest policy and encompasses today the less-studied social and political contexts. Whilst none of its species have been catalogued by the Spanish Government as invasive species, many local governments are assembling together to influence and reverse the situation. Taking this into account, this work studies the values, attitudes and political behaviors inherent in the fight against the expansion of eucalypts. Specifically, we examine at the local scale and through the cognitive hierarchy theory and discourse analysis, 20 cases of local governments that debate its own "Eucalypt-Free-Zone" declaration. Our analysis reveals that the conventional left-right spectrum is reproduced within the parties along their valuation of eucalypt plantations. Namely, those arguments based on the economic and social values adopt greater importance compared to the ecological and environmental ones. This hierarchy generates political isolation and motivates the absence of consensus. Therefore, we conclude that only an eventual decrease in wood price would motivate a relevant turn of political attitudes towards its management. Furthermore, the improvised inclusion of political debates perpetuates or even emphasizes the divergences of sectoral interests in the management of eucalypts.
[ "Discourse & Pragmatics", "Semantic Text Processing" ]
[ 71, 72 ]
SCOPUS_ID:77951908584
'Exergaming,' corporate interests and the crisis discourse of childhood obesity
In this paper we explore 'exergaming' in schools, the latest trend influencing the physical education (PE) curriculum. Throughout the paper we investigate the various proponents of exergaming including the gaming industry, academic researchers and PE teachers. Our inquiry takes a closer look at exergaming as a pedagogical trend and describes exergaming's potential implications for the profession of PE as well as for individual students. Our methodological approach includes review and analysis of popular, scholarly and curricular texts related to the use of exergaming in schools. The analysis underscores the political nature of the curriculum and, in so doing, invites a closer look at this curricular trend by examining whose interests are served through the circulation of crisis discourses surrounding youth (ill)health and the use of these crisis discourses to justify the placement of commercial products in schools. © 2010 Taylor & Francis.
[ "Discourse & Pragmatics", "Semantic Text Processing" ]
[ 71, 72 ]
SCOPUS_ID:85093112623
'Extimacy' (Extimité): From structural theory of language to affective theory of 'Ex-Centric' subject
The following exposure of the RSI topological complexities, orienting all the possible (inter)subjectivity, plays on the following two pairs of polarities: external/internal and linguistic/affective (it may be added: structure and topology). Lacan introduces the third possibility of human experience: “extimacy”, linking what is both excluded and intimate. The concept is the lacking link leading from structuralist approaches to language to thoroughly affective subjectivity of any speaking being. Spinosa's geometrical, highly dynamic system and his “differential calculus of affects” may account for the part that the vicissitudes of drive play in human existence as rooted in the deeply “extimate” sources.
[ "Emotion Analysis", "Sentiment Analysis", "Linguistics & Cognitive NLP", "Linguistic Theories" ]
[ 61, 78, 48, 57 ]
SCOPUS_ID:84863752749
'Feminism rules! Now, where's my swimsuit?' Re-evaluating feminist discourse in print media 1968-2008
Using both content and critical discourse analysis, this article traces the emergence of and changes in the ways feminism has been discursively constructed in 998 British and American news articles between 1968 and 1982 - which I define as the 'height' of the Second Feminist Wave, and 2008 - marking 40 years after feminism began gaining momentum in both nations. In analysing the British Times, Daily Mirror, Daily Mail, and Guardian newspapers, as well as the American New York Times, Washington Post, Chicago Tribune, Washington Times, I argue that not only has there been an erasure of feminist activism from these newspapers over time, but that discourses of feminism have become both de-politicized and de-radicalized since the 1960s, and can now largely be considered neoliberal in nature - a problematic construction for those seeking collective social change. © The Author(s) 2012.
[ "Discourse & Pragmatics", "Semantic Text Processing" ]
[ 71, 72 ]
SCOPUS_ID:52149093919
'Fire your proofreader!' Grammar correction in the writing classroom
This article critically reviews the usefulness of grammar correction in second language writing instruction through the eyes of five second-language writers. It first examines the validity of four teaching principles that appear to influence how writing instructors approach error correction in classrooms and concludes with discussions as to why grammar correction is necessary for second-language writers. © The Author 2007. Published by Oxford University Press; all rights reserved.
[ "Text Error Correction", "Syntactic Text Processing" ]
[ 26, 15 ]
SCOPUS_ID:84928728466
'Forty bucks is forty bucks': An analysis of a medical Doctor's professional identity
The study of narrative has focused on the narrator, often overlooking the transactional and relational role of the interlocutor, particularly in doctor-patient interactions where interactants co-construct the case. However, the role of the doctor as a narrative facilitator has rarely been explored. Using a case study from a doctor-patient interaction, this fine-grained discourse analysis demonstrates how a doctor assists his patient to construct her narrative while also enacting salient aspects of his own identity. The doctor uses discursive strategies, such as alignment, repair moves, and mitigation, which act as vehicles through which the doctor constructs his professional identity, providing 'space' for the patient's narrative, and assist in building trust and rapport.
[ "Discourse & Pragmatics", "Semantic Text Processing" ]
[ 71, 72 ]
SCOPUS_ID:80051511837
'Four years on, I'm ready to teach': Teacher education and the construction of teacher identities
This article discusses the results of a qualitative study that aimed to explore how one group of preservice English language teachers in Hong Kong constructed their identities as teachers. Using in-depth interviews to gain a rich understanding of participants' teacher identity formation in practice and discourse, the paper examines the perspectives of six preservice teachers about teaching and teachers at the completion of their undergraduate teacher education program. In contrast to the theorization of teacher identity construction, the results suggest that the participants often held rigid views about teaching and how they saw themselves, and others, as teachers. The paper argues that this rigidity may lead to antagonistic relations between these preservice teachers and their more experienced colleagues as the participants move into teaching and explores the implications for challenging this rigidity within the context of teacher education programs. Implications for future research are also considered. © 2011 Taylor & Francis.
[ "Discourse & Pragmatics", "Programming Languages in NLP", "Semantic Text Processing", "Multimodality" ]
[ 71, 55, 72, 74 ]
SCOPUS_ID:85137465848
'Frontrunners' Understanding of Universal Design in Architecture
In Denmark, the building sector is in a state of transition towards Universal Design (UD). Thus, UD has not yet completely found its way into the practice of architects and their clients. Legislation about accessibility has dominated. This paper studies understandings of UD through a discourse analysis based on a survey among professionals with experience and interest in UD and professionals who were expected to keep their fingers on the pulse of the profession's development. The findings illustrate the existence of five discourses: 1) Social sustainability, 2) Re-instatement of humans as a focal point, 3) It is not just about ramps, 4) Equality, and 5) Giving a voice. Across the discourses there exists a genuine attempt to legitimise and mainstream UD into the architectural practice, focusing on multisensory and architectural quality in the design of spaces for human diversity in all scales.
[ "Discourse & Pragmatics", "Semantic Text Processing" ]
[ 71, 72 ]
SCOPUS_ID:84876554317
'Full power despite stress': A discourse analytical examination of the interconnectedness of postfeminism and neoliberalism in the domain of work in an international women's magazine
Stories and images of successful career women and support for women's advancement in working life have become hallmarks of contemporary postfeminist media culture, and especially of women's magazines such as Cosmopolitan. While in previous research these features have been seen as signs for a new, popular feminism, more recently they have also been connected to the growing hegemony of neoliberal governance, a mode of power that ultimately aims at the economization of the social and is fundamentally exercised in and through discourse. The aim of this article is to investigate further the interconnectedness of these two phenomena, postfeminism and neoliberalism, in the domain of work, using the example of the German edition of Cosmopolitan. For a detailed and multilayered investigation the study draws on linguistically oriented discourse analysis, focusing on the operation of a 'discourse of postfeminist self-management'. The examination shows how this discourse, while on the one hand evoking an ethos of feminist engagement, on the other seeks to guide readers to mould themselves into a version of the entrepreneurial self required by the neoliberalized world of work. © The Author(s) 2013.
[ "Discourse & Pragmatics", "Semantic Text Processing" ]
[ 71, 72 ]
SCOPUS_ID:85052820768
'Funvjiefang' (Women's liberation) or 'Nvquanzhuyi' (feminism)? Transformation of China's national gender discourse in people's daily
Differentiating 'women's liberation' and 'feminism' provides a significant perspective to construe the women's movement in China. By employing discourse analysis of 1628 articles in People's Daily from 1949 to 2014, this paper presents that through certain discursive strategies Socialist Feminism is transformed into the hegemonic discourse of 'women's liberation' in China: 'women's liberation' is a socialist movement fighting against feudalism, capitalism, and imperialism under the leadership of the communist party; within the movement, gender equality is constructed mainly as obligation equality, and women have to work in the public sphere to gain their equal rights. Furthermore, women's liberation is not a stagnant discourse and has different focuses in different stages. Since the 1980s, the reappearance of 'feminism' discourse in China provides a new reference object to understand 'women's liberation'.
[ "Discourse & Pragmatics", "Semantic Text Processing" ]
[ 71, 72 ]
SCOPUS_ID:34248723216
'Getting behind the image': Personality politics in a Labour party election broadcast
The article examines a Labour party broadcast from the 1997 UK general election. I show how the film uses the conventions of a particular mode of contemporary documentary to present a portrait of Tony Blair. My main focus is on the way Blair's 'biography' is used for propaganda purposes, and how the tensions between the competing requirements of biography and propaganda manifest themselves textually. In particular, I examine Blair's strategic use of lifeworld discourses, and the role of pronominal choice in 'self' and 'other' referencing. Copyright © 2001 SAGE Publications.
[ "Discourse & Pragmatics", "Visual Data in NLP", "Semantic Text Processing", "Multimodality" ]
[ 71, 20, 72, 74 ]
SCOPUS_ID:85000909981
'Green Consumption' beyond mainstream economy: A discourse analysis
In contemporary society, green consumption is a popular concept. The life styles of people and consumption behaviors are moderated in accordance to the 'green ideology'. The process of green consumption can be observed through social behaviors such as preference of bio foods, recycling, reusing, limiting the over consumption and using environmentally friendly transport systems. However, mainstream economic analyses on green consumption argued that consumer behaviors are due to the rational choice of individuality based on utility and self- preferences. The hypothesis of this paper on consumer behavior in green consumption is configured by discourses according to the discourse analysis.
[ "Discourse & Pragmatics", "Semantic Text Processing" ]
[ 71, 72 ]
SCOPUS_ID:84921286083
'Greening the CAP' - Just a fashionable justification? A discourse analysis of the 2014-2020 CAP reform documents
Existing studies have employed discourse analysis to examine the reform of the EU's Common Agricultural Policy (CAP) but have only partially studied the implementation of the discourse in the context of CAP measures and budgetary distribution. The present study tries to fill this gap. By conducting a discourse analysis of the latest CAP reform (2014-2020) documents, we attempted to determine which discourses and discourse strategies predominated in the reform's documentation and how they were implemented into measures and budgetary distributions. The findings show that in the process of CAP reform decision-making, European institutions justified the CAP with a transformation of key discourses (productivist, multi-functional and neo-liberal) by emphasising the hugely popular environmental element while, at the same time, employing a strong productivist discourse at the level of measures and the budgetary distribution between the EU member states and farmers' groups. In order to retain a strong CAP as well as the current distribution of financial resources, the European Commission, the European Parliament and the European Council employed greening as a justification strategy as well as a productivist discourse as a major component in determining CAP measures.
[ "Discourse & Pragmatics", "Semantic Text Processing" ]
[ 71, 72 ]
SCOPUS_ID:85082552914
'Hakken en plakken': An exploration of the role of morphological awareness in spelling education
Orthography is considered to be a major problem in Dutch education, since many pupils don't seem to be able to master orthographic rules, even after years of education. In educational literature it is argued that the problems related to spelling are caused by approaches that focus more on rules of thumb than on linguistic insights. This is somewhat remarkable, since a good understanding of the Dutch orthographic system requires a fair amount of morphological knowledge. In order to effectively implement this knowledge, the development of a morphological awareness (MA) seems to be required. Therefore, a short intervention was designed for the upper levels of secondary schools (4 havo) which aimed to foster MA and, subsequently, improve orthographic skills. Results of this quasi-experimental study indicate that a short intervention can significantly boost MA, but that students don't seem to be able to use MA effectively to enhance spelling performance.
[ "Syntactic Text Processing", "Morphology" ]
[ 15, 73 ]
SCOPUS_ID:0034382994
'Heroin hell their own making': Construction of heroin users in the Australian press 1992-97
The ACT heroin trial was a proposal to evaluate the efficacy of prescription heroin as a treatment for heroin-dependent people. The trial was actively debated within the press by proponents and opponents but ultimately did not proceed due to a lack of required political support. Previous research indicates that public perceptions of the nature of drug users can influence the direction of policy responses. This paper analyses the construction of heroin users within press debate about the proposed ACT heroin trial, comparing and contrasting trial proponent and opponent views. The primary constructions of the user embraced models of users as people with health problems who were dying; who were criminals; classic deviance distinctions between us and them; that users posed costs and a threat to society; that users were victims; and discourse about 'ruined' selves. Despite attempts by trial proponents to construct the user as an 'ill us', the cultural value of abstinence from drug use and the ideology of individualism with its connotations of heroin use as a choice that required punishment rather than help were rarely challenged, reinforcing the view of drug use as a problem of individual morality.
[ "Discourse & Pragmatics", "Semantic Text Processing" ]
[ 71, 72 ]
SCOPUS_ID:84924370694
'Hey #311, come clean my street!': A spatio-temporal sentiment analysis of twitter data and 311 civil complaints
Twitter data has been applied to address a wide range of applications (e.g., Political election prediction and disease tracking), however, no studies have been conducted to explore the interactions and potential relationships between twitter data and social events available from government entities. In this paper, we introduce a novel approach to investigate the spatio-temporal relationships between the sentiment aspects of tweets and 311 civil complaints recorded in the 311 Case Database, which is freely available from the City of San Francisco. We also present results from two supporting tasks: (1) We apply sentiment analysis techniques to model the emotional characteristics of five metropolitan areas around the globe, allowing one to gain insight into the relative happiness across cities and neighborhoods within a city, and (2) we quantify the performance of several open-source machine learning algorithms for sentiment analysis by applying them to large volume of twitter data, thereby providing empirical guidelines for practitioners. Major contributions and findings include (1) We have developed a system for the relative ranking of happiness of a geographical area. Our results show that Sydney, Australia is the happiest of the five cities under study, (2) We have found a counterintuitive positive correlation between 311-report frequency and local sentiment, and (3) When performing sentiment analysis of tweets, the inclusion of emoticons in the training dataset can lead to model over fitting, whereas NLP-based features seem to have a great potential to improve the classification accuracy.
[ "Sentiment Analysis" ]
[ 78 ]
SCOPUS_ID:34247395041
'Honey, I'm home!': Framing in family dinnertime homecomings
Dinnertime has served as a lucrative site for the study of family discourse (e.g., Blum-Kulka 1997; Ochs and Taylor 1995). Its importance as a site of analysis is not surprising since dinner is one of the few activities that brings many families together on a daily basis and, as such, serves as an important site for the constitution and maintenance of the family and familial roles. In this study of the naturally occurring interaction of dual-income families with young children, the family members do not eat dinner at the same time. To understand family discourse at mealtime in these families, it is necessary to redefine 'family dinner' to include meals in which all family members are present but not everyone is eating, and to consider the wider context in which these interactions occur; that is, as part of homecoming transitions when one parent arrives home and the child and other parent are already there. I use a framing approach (Goman 1981; Tannen 1994) to consider two 'dinnertime homecomings' from two dual-earner families who tape-recorded themselves from morning until night for at least one week. I find four primary frames that occur during these interactions (dinner, managerial, caregiving, sociable) and focus on one footing that occurs within the sociable frame: a child-centered family footing. I describe how this footing is linguistically constituted during the interactions, how parents negotiate this footing as it competes with other footings and frames, and how one parent uses this footing as a resource to restore family harmony. Ultimately, the child-centered footing in a sociable frame is a family-centered footing that serves to create and maintain family solidarity. © Walter de Gruyter 2006.
[ "Discourse & Pragmatics", "Semantic Text Processing" ]
[ 71, 72 ]
SCOPUS_ID:84905243588
'Houston, We have a solution' : Using NASA Apollo program to advance speech and language processing technology
NASA's Apollo program stands as one of mankind's greatest achievements in the 20th century. During a span of 4 years (from 1968 to 1972), a total of 9 lunar missions were launched and 12 astronauts walked on the surface of the moon. It was one the most complex operations executed from scientific, technological and operational perspectives. In this paper, we describe our recent efforts in gathering and organizing the Apollo program data. It is important to note that the audio content captured during the 7-10 day missions represent the coordinated efforts of hundreds of individuals within NASA Mission Control, resulting in well over 100k hours of data for the entire program. It is our intention to make the material stemming from this effort available to the research community to further research advancements in speech and language processing. Particularly, we describe the speech and text aspects of the Apollo data while pointing out its applicability to several classical speech processing and natural language processing problems such as audio processing, speech and speaker recognition, information retrieval, document linking and a range of other processing tasks which enable knowledge search, retrieval, and understanding.. We also highlight some of the outstanding opportunities and challenges associated with this dataset. Finally, we also present initial results for speech recognition, document linking, and audio processing systems. Copyright © 2013 ISCA.
[ "Programming Languages in NLP", "Speech & Audio in NLP", "Information Retrieval", "Multimodality" ]
[ 55, 70, 24, 74 ]
SCOPUS_ID:85126772366
'How Robust R U?': Evaluating Task-Oriented Dialogue Systems on Spoken Conversations
Most prior work in dialogue modeling has been on written conversations mostly because of existing data sets. However, written dialogues are not sufficient to fully capture the nature of spoken conversations as well as the potential speech recognition errors in practical spoken dialogue systems. This work presents a new benchmark on spoken task-oriented conversations, which is intended to study multi-domain dialogue state tracking and knowledge-grounded dialogue modeling. We report that the existing state-of-the-art models trained on written conversations are not performing well on our spoken data, as expected. Furthermore, we observe improvements in task performances when leveraging $n$-best speech recognition hypotheses such as by combining predictions based on individual hypotheses. Our data set enables speech-based benchmarking of task-oriented dialogue systems.
[ "Speech & Audio in NLP", "Robustness in NLP", "Natural Language Interfaces", "Text Generation", "Dialogue Systems & Conversational Agents", "Responsible & Trustworthy NLP", "Speech Recognition", "Multimodality" ]
[ 70, 58, 11, 47, 38, 4, 10, 74 ]
SCOPUS_ID:0033405743
'How am I gonna answer this one?': A discourse analysis of fathers' accounts of providing sexuality education for young sons
This paper presents a discourse analysis of fathers' accounts of their communication about sexuality with their young sons. In-depth interviews were conducted with eight Australian men who have six-year-old sons. Participants were asked to describe their experiences of and feelings about how they learned about sexuality and how they teach their sons. The transcripts of these interviews were analyzed using the Textually Oriented Discourse Analysis described by Fairclough. Seven key assumptions which underlie the participants' discourse are identified. While these assumptions present significant obstacles for many fathers, several are actively struggling with these dominant ideologies. Sexuality educators may be able to help address this conflict by encouraging fathers to engage in discussion about their fears and beliefs about sexuality education. This research suggests that many fathers need assistance to participate in this task.
[ "Discourse & Pragmatics", "Semantic Text Processing" ]
[ 71, 72 ]
SCOPUS_ID:84902136531
'How do I do it if I don't like writing?': Adolescents' stances toward writing across disciplines
This research embedded in the National Study of Writing Instruction examines higher- and lower-achieving adolescents' stances toward content-area writing through a qualitative discourse analysis of interviews with 40 students in California, Kentucky, New York, and Texas secondary schools. The study asked: (1) How do students' stances toward writing compare in general and across disciplines? (2) How do stances compare among middle and high school students and among students with different achievement histories? Results suggest that adolescents generally hold positive attitudes toward writing that allows for the expression of subjective stances, which they report is more commonly assigned in English language arts classrooms. Implications for the adoption of new US standards for disciplinary writing are discussed. © 2013 Springer Science+Business Media Dordrecht.
[ "Discourse & Pragmatics", "Semantic Text Processing" ]
[ 71, 72 ]
SCOPUS_ID:0034389460
'I Am Not a Patient, and I Am Not a Child': The Institutionalization and Experience of Pregnancy
In this article the focus is on how the relation between the self and body is formulated in medical/healthcare discourses and how these affect the experiences of pregnant women. I draw on data collected during research on the self-image of young mothers, analyses of booklets and handouts distributed to pregnant women, and interviews conducted both on individual and group bases with young mothers. I argue that the normalizing tendencies identified in the booklets strip women of their agency. However, pregnant women do not always position themselves in terms of maternal normativities. Their accounts of pregnancy and childbirth both support and challenge the knowledge that underpins the practices of medical/healthcare institutions. Their position as agents matters a great deal for them and affects the extent to which they experience pregnancy and childbirth positively or negatively.
[ "Discourse & Pragmatics", "Semantic Text Processing" ]
[ 71, 72 ]
SCOPUS_ID:84935083028
'I Can't Go Back Because If I Go Back I Would Die': How Asylum Seekers Manage Talk about Returning Home by Highlighting the Importance of Safety
Asylum seekers living in the UK have been shown to have fled danger in their countries of origin, only to face hardship and the threat of deportation once there. This paper draws on the discursive psychological approach to address the way in which asylum seekers in the UK manage questions about returning to their country of origin. Interviews were conducted with nine asylum seekers in a refugee support centre in the Midlands. The interviews were transcribed and analysed using discourse analysis. The analysis showed that participants drew on the notion of safety to counter suggestions that they should return to their country of origin and to manage their identity as legitimate asylum seekers in need of support. The use of this strategy and the use of interviews for discursive analysis are discussed.
[ "Discourse & Pragmatics", "Semantic Text Processing" ]
[ 71, 72 ]
SCOPUS_ID:85124712467
'I Let Depression and Anxiety Drown Me...': Identifying Factors Associated With Resilience Based on Journaling Using Machine Learning and Thematic Analysis
Over the years, there has been a global increase in the use of technology to deliver interventions for health and wellness, such as improving people's mental health and resilience. An example of such technology is the Q-Life app which aims to improve people's resilience to stress and adverse life events through various coping mechanisms, including journaling. Using a combination of sentiment analysis and thematic analysis methods, this paper presents the results of analyzing 6023 journal entries from 755 users. We uncover both positive and negative factors that are associated with resilience. First, we apply two lexicon-based and eight machine learning (ML) techniques to classify journal entries into positive or negative sentiment polarity, and then compare the performance of these classifiers to determine the best performing classifier overall. Our results show that Support Vector Machine (SVM) is the best classifier overall, outperforming other ML classifiers and lexicon-based classifiers with a high F1-score of 89.7%. Second, we conduct thematic analysis of negative and positive journal entries to identify themes representing factors associated with resilience either negatively or positively, and to determine various coping mechanisms. Our findings reveal 14 negative themes such as stress, worry, loneliness, lack of motivation, sickness, relationship issues, as well as depression and anxiety. Also, 13 positive themes emerged including self-efficacy, gratitude, socialization, progression, relaxation, and physical activity. Seven (7) coping mechanisms are also identified including time management, quality sleep, and mindfulness. Finally, we reflect on our findings and suggest technological interventions that address the negative factors to promote resilience.
[ "Information Extraction & Text Mining", "Information Retrieval", "Text Classification", "Sentiment Analysis" ]
[ 3, 24, 36, 78 ]
SCOPUS_ID:85104266739
'I Tried to Breastfeed but...': Exploring Factors Influencing Breastfeeding Behaviours Based on Tweets Using Machine Learning and Thematic Analysis
Social media is a growing platform for health-related discourse, opinion and experience sharing, including breastfeeding. For instance, nursing mothers share their personal experiences and opinions about breastfeeding on social networks, such as Facebook and Twitter. Unravelling the sentiments behind these experiences will promote adequate knowledge of many challenges, benefits, and factors influencing breastfeeding behaviours. To achieve this, we mine breastfeeding-related tweets and then perform sentiment analysis of the tweets using lexicon-based and machine learning (ML) techniques with the aim of detecting their sentiment polarity (i.e., positive or negative). Specifically, we implement and compare four lexicon-based sentiment classifiers, as well as five ML-based classifiers. Our results show that VADER-EXT (our extended version of VADER) performed best with an overall F1-score of 82.4%, compared to the other lexicon-based classifiers. On the other hand, Support Vector Machine (SVM) outperformed the other four ML-based classifiers with an overall F1-score of 73.7%. The overall best performing classifier is then used in determining the sentiment polarity of tweets. Next, we conduct thematic analysis of both positive and negative tweets to identify the factors influencing breastfeeding behaviours either positively or negatively. Our findings reveal various health-related factors (such as lactational issues, medical issues, and nutritional issues), social factors, psychological factors, and situational factors affecting breastfeeding behaviours negatively. Also, perceived benefits, maternal self-efficacy, social support, and education and training support emerged as the positive factors influencing breastfeeding behaviours. Finally, we reflect on our findings and recommend interventions that address the negative factors to promote positive breastfeeding behaviours.
[ "Text Classification", "Polarity Analysis", "Sentiment Analysis", "Information Retrieval", "Information Extraction & Text Mining" ]
[ 36, 33, 78, 24, 3 ]
SCOPUS_ID:84866283911
'I am a man!' the daily sun campaign and gender violence
The 'Charter for a Man' campaign was run in the Daily Sun, the widely circulated South African tabloid newspaper, from 7 November to 7 December 2007. The campaign, ostensibly designed to discourage gender violence, could potentially provide social critique and be transformative of gender inequalities. This critical investigation of the campaign coverage sets out to establish the gendered discourse and the forms of masculinities and femininities privileged therein. Informed by poststructural feminist thought and an eclectic discourse analysis approach, it argues that the Daily Sun defined itself as the self-declared hero - a position endorsed by the celebrity signatories they chose to include. If 'good' men were constituted as powerful and chivalrous in contrast to 'bad', abusive men, women were present merely as victims, and in a single case as a celebrity who rehearsed a patriarchal discourse of men as powerful and women as needy. A strong theme of nationhood, particularly in the editorials, endorsed the identity of men as manly, black (South) Africans. While the campaign held the promise of a progressive initiative, it endorsed the existing gender order through the repeated representation of women as weak and in need of patronage, and men as their protectors and providers. © 2012 Institute for Media Analysis in South Africa.
[ "Discourse & Pragmatics", "Semantic Text Processing" ]
[ 71, 72 ]
SCOPUS_ID:77951736629
'I connected so well with it': A teen mother talks about reading
In this paper, one 19-year-old married mother who was sexually abused when younger discusses the painful and important role reading has played in her life. By using narrative discourse analysis, I parse the young mother's spoken narrative and illuminate the emotions that cluster around her reading experiences. I discover that she has used reading as a way to make sense of the traumatic experiences of her childhood. I discover as well that, though she evidences immaturity, her moral thinking - refined through her roles as daughter, mother, and wife - is more nuanced than are the texts she is able to read. The questions I ask about her relationship to these books can be generalized to many non-traditional students going to high schools, or reading in book clubs, today. © 2010 Taylor & Francis.
[ "Discourse & Pragmatics", "Semantic Text Processing" ]
[ 71, 72 ]
SCOPUS_ID:79955701963
'I don't f***ing care!' marginalia and the (textual) negotiation of an academic identity by university students
This article charts the ways in which students negotiate an academic identity whilst pursuing academic tasks that are publicly observable precisely as 'academic tasks' to their peers. Previous research into aspects of student interaction that take place within university tutorial sessions has suggested that different kinds of student identity come into conflict as students interact, face-to-face. Most notably, the imperative of 'doing education' - as a keen proto-academic seeking a good final degree classification - is often overridden by the imperative of 'doing being a student' - as an average and/or indifferent student who does not stand out. In this article, however, it is not the university tutorial but a slightly different and hitherto unexplored site of interaction that is foregrounded: the university library textbook. As with tutorials, the margins of library textbooks are spaces within which individually accomplished research becomes publicly observable - in the former as group work, and in the latter as student marginalia. Drawing on discourse analytic concepts such as 'footing' and 'stake' this article seeks to understand what student marginalia can tell us about the ways in which the tension between 'doing education' and 'doing being a student' is managed and negotiated in sites other than the university tutorial. © The Author(s) 2011.
[ "Discourse & Pragmatics", "Semantic Text Processing" ]
[ 71, 72 ]
SCOPUS_ID:84901237943
'I don't want to see my children suffer after birth': The 'risk of knowing' talk and decision-making in prenatal screening for Down's syndrome in Hong Kong
In this article, we examine the 'risk of knowing' talk (Sarangi, S., et al., 2003. 'Relatively speaking': relativisation of genetic risk in counselling for predictive testing. Health, risk & society, 5 (2), 155-170, p. 155) in prenatal screening for Down's syndrome in Hong Kong. The 'risk of knowing' talk refers to the consequences of learning about a health condition, such as the psychosocial and interpersonal implications of testing, and the subsequent management of the condition. The stigma of eugenics and that the termination of pregnancy is the only available 'medical intervention' imply that the risk talk and decision-making in prenatal screening carry serious ethical, moral and social implications (Pilnick, A. and Zayts, O.A, 2012. 'Let's have it tested first': choice and circumstances in decision-making following positive antenatal screening in Hong Kong. Sociology of health and illness, 34 (2), 266-282). This issue has not attracted much attention in the previous literature. This study is part of a larger project on prenatal screening conducted in one Prenatal Diagnostics and Counselling Department of a Hong Kong hospital in 2006-2013. It draws on 20 video-recorded consultations with pregnant women who had received a 'positive' (high risk) screening result and were invited to consider further diagnostic testing. Using theme-oriented discourse analysis (Roberts, C. and Sarangi, S., 2005. Theme-oriented discourse analysis of medical encounters. Medical education, 39 (6), 632-640), we show that in these consultations, the 'risk of knowing' talk was not initiated by the health care professionals. It might, however, be evoked by the women. We examine the impact of the 'risk of knowing' on decision-making, and discuss specific discourse (linguistic and rhetorical) devices that the participants employed to negotiate three competing agendas: the health care professionals' preference of diagnostic testing, clients' concerns of having a baby with Down's syndrome and the overarching professional goal of these encounters of facilitating the clients' informed choice regarding further testing. © 2014 © 2014 Taylor & Francis.
[ "Discourse & Pragmatics", "Semantic Text Processing" ]
[ 71, 72 ]
SCOPUS_ID:84863546844
'I just couldn't do it': Representations of constraint in an oral history corpus
Corpus linguistic techniques are increasingly being used by discourse analysts whose interest is in the 'critical' issues of inequality and the representation of disadvantaged groups. This paper reports an extension of these approaches, where concordancing was used to analyse a corpus of 144 transcribed oral history interviews in order to explore the issue of constraint on the speakers' goals and experiences. The analysis is of the expression I couldn't, which is contextualised with reference to research on negation and modality in authentic discourse contexts. This paper explores the ways in which I couldn't is deployed to refer to constraints of three main kinds: physical (pertaining to the body and material objects), structural (pertaining to the distribution of resources) and cultural (pertaining to social norms and expectations). The approach illustrates the advantage of maintaining an analytical distinction between the discursive and the material, so as to explore the interplay between them. © 2012 Copyright Taylor and Francis Group, LLC.
[ "Discourse & Pragmatics", "Semantic Text Processing", "Representation Learning" ]
[ 71, 72, 12 ]
SCOPUS_ID:84875647826
'I just want to be normal': An analysis of discourses of normality among recovering heroin users
Research that has explored the lives of men and women recovering from heroin addiction has reported that users often claim that they 'just want to be normal'. Working within a Foucauldian tradition, we argue in this article that the notions of 'governmentality' and the 'norm' are especially apposite to understanding the ubiquity of this aspiration. Here we focus not on the formal institutions of governance that encourage individuals to adhere to social, cultural and political norms, but rather seek to explore recovering users' accounts of normality as they are envisaged and expressed. The reported empirical data were generated from interviews with 40 men and women in England at various stages of recovery from heroin use. The analytic focus is upon the accounts of normality articulated during the interviews in order to identify the ways in which being normal is presented by the participants. In keeping with the methodological tradition of discourse analysis we identify six discursive repertoires of 'normality talk' that transcend the accounts. It is concluded that the negotiation of normality is a precarious route for this social group. Articulations of a desire to be normal are replete with tensions; there are expressions of both resistance and resignation. Despite claims by some contemporary social theorists that diversity is the 'new normality', the accepted bounds of 'difference' are limited for those who have been addicted to heroin. © The Author(s) 2012.
[ "Discourse & Pragmatics", "Semantic Text Processing" ]
[ 71, 72 ]
SCOPUS_ID:84940960642
'I really needed help': What mothers say about their post-birth care in Queensland, Australia
Background: Australian mothers consistently rate postnatal care as the poorest aspect of their maternity care, and researchers and policymakers have widely acknowledged the need for improvement in how postnatal care is provided. Aim: To identify and analyse mothers' comments about postnatal care in their free text responses to an open ended question in the Having a Baby in Queensland Survey, 2010, and reflect on their implications for midwifery practice and maternity service policies. Methods: The survey assessed mothers' experiences of maternity care four months after birth. We analysed free-text data from an open-ended question inviting respondents to write 'anything else you would like to tell us'. Of the final survey sample (N= 7193), 60% (N= 4310) provided comments, 26% (N= 1100) of which pertained to postnatal care. Analysis included the coding and enumeration of issues to identify the most common problems commented on by mothers. Comments were categorised according to whether they related to in-hospital or post-discharge care, and whether they were reported by women birthing in public or private birthing facilities. Results: The analysis revealed important differences in maternal experiences according to birthing sector: mothers birthing in public facilities were more likely to raise concerns about the quality and/or duration of their in-hospital stay than those in private facilities. Conversely, mothers who gave birth in private facilities were more likely to raise concerns about inadequate post-discharge care. Regardless of birthing sector, however, a substantial proportion of all mothers spontaneously raised concerns about their experiences of inadequate and/or inconsistent breastfeeding support. Conclusion: Women who birth in private facilities were more likely to spontaneously report concerns about their level of post-discharge care than women from public facilities in Queensland, and publically provided community based care is not sufficient to meet women's needs. Inadequate or inconsistent professional breastfeeding support remains a major issue for early parenting women regardless of birthing sector.
[ "Ethical NLP", "Responsible & Trustworthy NLP" ]
[ 17, 4 ]
SCOPUS_ID:85102946940
'I understand'-initiated formulations of the other: A semi-fixed claim to the intersubjective
Some language patterns appear fixed at a certain time, enabling their description as grammatical structures. Semi-fixed patterns that routinely accomplish specific social actions constitute more of an analytical challenge. This chapter targets the phrase ma saan aru 'I understand' in Estonian together with the ensuing other-attentive formulation '2nd person expression + a cognitive concept' and argues that it is a semi-fixed expression, a "claim to the intersubjective", that manages a misalignment between participants. While claiming to have successfully accessed the other's motives or feelings, the speaker regularly advances her own agenda through the formulation of the other. This suggests a systematic relationship between cognitive lexicon, grammatical structure, and interactional function, and calls for a language theory that incorporates semi-fixedness.
[ "Linguistics & Cognitive NLP", "Linguistic Theories" ]
[ 48, 57 ]
SCOPUS_ID:34248740681
'I want to be a prime minister', or what linguistic choice can do for campaigning politicians
The aim of this article is to analyse the strategies that politicians can use in order to defeat their political adversaries. To this end, I have put some ideas developed by discourse analysts to the test, taping some of the speeches, interviews and debates of the 2000 electoral campaign in Andalusia (Spain), scrutinizing the four main candidates' most significant discursive devices and paying special attention to the way they interact with each other, their interviewers and the audience in their political meetings. In this way, I have tried to see whether their different political persuasions may produce a characteristically different distinctive linguistic style, whether gender might influence their choice of discourse structures, and the extent to which the winning candidate's linguistic idiosyncrasies might have contributed to his success.
[ "Discourse & Pragmatics", "Semantic Text Processing" ]
[ 71, 72 ]
SCOPUS_ID:84890807762
'I'M NOT HAPPY, BUT I'M OK': How asylum seekers manage talk about difficulties in their host country
This paper addresses the ways in which asylum seekers in the UK manage making complaints about their host country. The authors demonstrate that asylum seekers have fled dangerous situations in their countries of origin and then can face difficulties and hostility in the UK. A discursive psychological approach is used to assess the ways in which asylum seekers made complaints regarding their treatment. Interviews were conducted in a refugee centre in the Midlands with nine asylum seekers and were transcribed for a discourse analysis to be conducted. Analysis of the data showed that participants criticised the asylum system for being unfair. They also made claims about not being happy in the UK, but did so in ways that downgraded the problem so as to manage the possible dilemma of appearing ungrateful and undermining their reasons for claiming asylum. The problems associated with these strategies are discussed. © 2013 Taylor & Francis.
[ "Discourse & Pragmatics", "Semantic Text Processing" ]
[ 71, 72 ]
SCOPUS_ID:84905450342
'I'm an expert in me and I know what I can cope with': Patient expertise in rheumatoid arthritis
The active involvement of patients in decisionmaking and the focus on patient expertise in managing chronic illness constitutes a priority in many healthcare systems including the NHS in the UK. With easier access to health information, patients are almost expected to be (or present self) as an 'expert patient' (Ziebland 2004). This paper draws on the meta-analysis of interview data collected for identifying treatment outcomes important to patients with rheumatoid arthritis (RA). Taking a discourse approach to identity, the discussion focuses on the resources used in the negotiation and coconstruction of expert identities, including domainspecific knowledge, access to institutional resources, and ability to self-manage. The analysis shows that expertise is both projected (institutionally sanctioned) and claimed by the patient (self-defined). We close the paper by highlighting the limitations of our pilot study and suggest avenues for further research. Copyright © Equinox Publishing Ltd.
[ "Discourse & Pragmatics", "Semantic Text Processing" ]
[ 71, 72 ]
SCOPUS_ID:0038225055
'I'm just saying...': Discourse markers of standpoint continuity
Examining discourse markers (Schiffrin, 1987) in two transcribed discussions of controversial issues in an undergraduate 'critical thinking' class, we note frequent uses of 'I'm just saying' and related metadiscursive expressions (I'm/we're saying, I'm/we're not saying, etc.). Our central claim is that these 'saying' expressions are pragmatic devices by which speakers claim 'all along' to have held a consistent argumentative standpoint, one that continues through the discussion unless changed for good reasons. Through close analysis of a series of discourse examples, we show how these discourse markers are used to display continuity, deflect counterarguments, and acknowledge the force of counterarguments while preserving continuity. In a concluding section we reflect critically on the use of these continuity markers with regard to four pragmatic functions that they potentially serve: to specify and clarify argumentative standpoints, to acknowledge a presumption of standpoint continuity, to acknowledge a normative expectation that discussion participants should have standpoints, and to avoid overt disagreement while saving face. © 2000 Kluwer Academic Publishers.
[ "Discourse & Pragmatics", "Semantic Text Processing" ]
[ 71, 72 ]