Datasets:

TimSchopf

/

nlp_taxonomy_data

like 0

SCOPUS_ID:85111628502

"did you buy it already", Detecting Users Purchase-State from Their Product-Related Questions

In this study we address the problem of identifying the purchase-state of users, based on product-related questions they ask on an eCommerce website. We differentiate between questions asked before buying a product (pre-purchase) and after (post-purchase). At first, we study the ambiguity that exists in purchase-states' definition, and then investigate the linguistic characteristics of the questions in each state. We analyze the discrepancy between the language models of pre- and post-purchase questions, and offer two classification schemes for this task, both outperform human judgments. We additionally show the effectiveness of our classification models in improving real world applications for both consumers and sellers.

SCOPUS_ID:84989821863

"doctor smartphone": A dispositive analysis of the norwegian press's presentation of m-health applications

The rapid growth in the field of m-health has not gone unnoticed by the mainstream media in Norway. Norwegian newspapers have a strong presence and outreach and hence play an important role in shaping of the public discourse on various subjects with m-health being no exception. This article presents a Dispositive Analysis of 23 articles from 6 national newspapers concerning mobile health applications. The analysis resulted in an interpretation of the press's technology views as theories of technology, which informed the discussion in this paper. Further, the newspaper articles were understood as discursive practices and analyzed by applying the concept of dispositives. The results of the analysis suggest inclusion of Dispositive Analysis as a step in Participatory Design process as means of enriching the design practices as well as uncovering the marginalized 'voices' and thus addressing the call for democratization of technology.

SCOPUS_ID:85090234961

"elinor Is talking to me on the screen!" integrating conversational agents into children's television programming

Science-oriented television and video programming can be an important source of science learning for young children. However, the educational benefits of television have long been limited by children not being able to interact with the content in a contingent way. This project leverages an intelligent conversational agent -an on-screen character capable of verbal interaction-to add social contingency into children's experience watching science videos. This conversational agent has been developed in an iterative process and embedded in a new PBS KIDS science show "Elinor Wonders Why." This Late Breaking Work presents the design of the conversational agent and reports findings from a field study that has proven feasibility of this approach. We also discuss our planned future work to examine the agent's effectiveness in enhancing children's engagement and learning.

SCOPUS_ID:84954242417

"ethereal Carbon": Legitimizing liquefied natural gas in British Columbia

This paper examines the provincial government of British Columbia's recent proposal of building a Liquefied Natural Gas (LNG) industry, in which natural gas extracted through hydraulic fracturing will be liquefied and then exported to Asian markets. Drawing upon the growing literature on energopower, petro-state, and petro-culture, selected texts and images from "LNG in BC" - the project's official branding website - were analyzed through a multimodal critical discourse analysis. The results reveal two primary strategies of legitimation: the first emphasizes the economic benefits of LNG development in terms of employment and taxation revenues; the second defines LNG as a means of strengthening B.C.s environmental leadership. The second narrative depends heavily upon the symbolic values of natural gas, contrasting its "clean" appearance (as a colorless and odorless gas) with the material density and "toxic sensuality" of other "dirty" fossil fuels (such as coal, oil, and bitumen). The website also presents a linear and simplified "storyline" of the generation of LNG which emphasizes the simple, "clean" process of liquefaction to distract attention from the ecological and health risks of hydraulic fracturing.

SCOPUS_ID:85109338184

"geographical surprises" and more: The poetics of s.n. durylin s vodlozero diary

The article analyses the poetics of S.N. Durylin s unpublished North (Vodlozero) diary. The researcher of this diary has the following tasks: To study the text from a functional point of view, to consider the features of genre form and genre content, to try to recreate the image of the author based on the text of the diary. Functionally, Durylin s Northern diary is interesting because it is a travel essay about the North, in which the author describes not only historically significant landscapes, but also customs of the northerners. The author reveals the intimate experience of what he saw in the North, while the text is distinguished by the objectification of the described, since the notes were supposed to be handed over to the Archaeological Institute. In this case, for understanding the content of the diary genre, it is very important that this is Durylin s first diary, created on the basis of daily records in July August 1917, the diary of the so-called formation period, which caused another of its functions "spiritual chronograph". Thus, Durylin s Vodlozero diary recreates the process of spiritual formation of the writer s personality, and is also a valuable artistic evidence of the Russian national picture of the world.

SCOPUS_ID:85094215915

"i Hope This Is Helpful": Understanding Crowdworkers' Challenges and Motivations for an Image Description Task

AI image captioning challenges encourage broad participation in designing algorithms that automatically create captions for a variety of images and users. To create large datasets necessary for these challenges, researchers typically employ a shared crowdsourcing task design for image captioning. This paper discusses findings from our thematic analysis of 1,064 comments left by Amazon Mechanical Turk workers using this task design to create captions for images taken by people who are blind. Workers discussed difficulties in understanding how to complete this task, provided suggestions of how to improve the task, gave explanations or clarifications about their work, and described why they found this particular task rewarding or interesting. Our analysis provides insights both into this particular genre of task as well as broader considerations for how to employ crowdsourcing to generate large datasets for developing AI algorithms.

SCOPUS_ID:85062280078

"i am I": Self-constructed transgender identities in internet-mediated forum communication

This article analyses identity constructions and representations of self-identifying transgender individuals on a web-based forum. Although the forum is aimed towards all transgender users, the primary user-group are transfeminine users (intending on) undergoing medico-surgical interventions to align their physiology and identity. The data for this analysis are initial text posts from the forum board used for introductions (i.e. new users of the forum introducing themselves). The article assumes that introductions are the context in which one asserts key identity features; hence, this board is the most pertinent for analysing identity construction. In this article, I use a combination of corpus linguistics and Critical Discourse Studies tools to analyse the use of pronouns and gender-indexical nouns in identity constructions and the representation of social categorisations. This article is an attempt to demonstrate that transgender is not a collective homogeneous identity, and that gender-sex incongruence may not be a salient identity feature for some forum-users. I also examine the ideologies (re)produced in the local forum-communication discourse, and the evaluation of hegemonic practices within transgender discourse and wider gender discourse to further demonstrate the heterogeneity of transgender identity.

SCOPUS_ID:84957703097

"i am a cheerleader, but secretly i deal drugs" authenticity through concealment and disclosure

Authenticity requires a balance between being true to others and being true to the self. People keep and disclose secrets in order to maintain authenticity contextually within relationships, as well as across contexts through self-reflexive evaluations. Based upon a content and discourse analysis of 1600 submissions to the PostSecret mail-art project, this study explores how secrets are used to manage disparate social, role, and personal identities to sustain coherent selves in some contexts and incoherent selves in others. Secrets are revealed anonymously through the PostSecret mail-art project, which allows for a cathartic release without disrupting their performances in front of significant others. This project analyzes the attempts made to maintain or achieve authenticity through the dialectical acts of concealment and disclosure.

SCOPUS_ID:85090192247

"i can feel your empathic voice": Effects of nonverbal vocal cues in voice user interface

Active use of voice assistant highlights the importance of interpersonal relationships with the conversational voice agent. When the agent evaluates a user's task performance, empathic feedback is needed to prevent the user's negative feelings. Voice reflects emotions through its nonverbal voice features that can either strengthen or lessen the feeling of empathy. Our study investigated the effect of nonverbal vocal cues in speech interaction on the user's perception toward the agent. 39 university students participated in the experiment, and MANOVA was tested to analyze their responses regarding intimacy, similarity, connectedness, enjoyment, and ease of use. The study result showed that using nonverbal vocal cues on empathic feedback contributes to establishing an interpersonal relationship with the agent, which gives implications to the fields of human-centered agent design.

SCOPUS_ID:52149106039

"i know what you feel": Analyzing the role of conjunctions in automatic sentiment analysis

We are interested in finding how people feel about certain topics. This could be considered as a task of classifying the sentiment: sentiment could be positive, negative or neutral. In this paper, we examine the problem of automatic sentiment analysis at sentence level. We observe that sentence structure has a fair contribution towards sentiment determination, and conjunctions play a major role in defining the sentence structure. Our assumption is that in presence of conjunctions, not all phrases have equal contribution towards overall sentiment. We compile a set of conjunction rules to determine relevant phrases for sentiment analysis. Our approach is a representation of the idea to use linguistic resources at phrase level for the analysis at sentence level. We incorporate our approach with support vector machines to conclude that linguistic analysis plays a significant role in sentiment determination. Finally, we verify our results on movie, car and book reviews. © 2008 Springer-Verlag Berlin Heidelberg.

SCOPUS_ID:84863275820

"i loan because.": Understanding motivations for pro-social lending

As a new paradigm of online communities, microfinance sites such as Kiva.org have attracted much public attention. To understand lender motivations on Kiva, we classify the lenders' self-stated motivations into ten categories with human coders and machine learning based classifiers. We employ text classifiers using lexical features, along with social features based on lender activity information on Kiva, to predict the categories of lender motivation statements. Although the task appears to be much more challenging than traditional topic-based categorization, our classifiers can achieve high precision in most categories. Using the results of this classification along with Kiva teams information, we predict lending activity from lender motivation and team affiliations. Finally, we make design recommendations regarding Kiva practices which might increase pro-social lending. Copyright 2012 ACM.

SCOPUS_ID:84921790032

"i will stop your mouth": The regulation of jesting in Much Ado about Nothing

Beatrice's outrage at Benedick's suggestion that she cribs her "good wit out of The Hundred Merry Tales" (2.1.119-20) gives a starting point for an analysis of the function-and the regulation-of witty speech in Much Ado About Nothing. Using concepts drawn both from linguistic pragmatics and classical rhetoric, this essay considers the very specific social functions of laughter in Beatrice's speech, as contrasted to that of jest-book speakers, Margaret and Hero. This reading reveals the extent to which Beatrice actually withholds her wit, and reveals a largely unremarked strain of social aggression in the speech of Hero.

SCOPUS_ID:84880115641

"i" - A novel algorithm for optical character recognition (OCR)

Computer vision, artificial intelligence and pattern recognition have been important areas of research for a while in the history of electronics and image processing. Optical character recognition (OCR) is one of the main aspects of computer vision and has evolved greatly since its inception. OCR is a method in which readable characters are recognized from optical data obtained digitally. Many methodologies and algorithms have been developed for this purpose using different approaches. Here we present one such approach for OCR named " i ". Amongst all other OCR systems available, " i " aims at a high speed, simple, font independent and size independent OCR system based on a unique segment extraction technique. This algorithm can be used as a kernel for single alphabet detection within a complete OCR solution system without the need for any complex mathematical operations. The highlight of this methodology is that, it does not use any libraries or databases of image matrices to recognize alphabets, but it has a unique algorithm to recognize alphabets instead. This algorithm has been implemented in MATLAB 7.14.0.739 build R2012a on a test set of 500 images of text and an accuracy of 100% for three font families namely Arial, Times New Roman and Courier New has been obtained. © 2013 IEEE.

SCOPUS_ID:85028707872

"i'm an Addict" and other sensemaking devices: A discourse analysis of self-reflections on lived experience of social media

How do young people make sense of their social media experiences, which rhetoric do they use, which grand narratives of technology and social media do they rely on? Based on discourse analysis of approximately 500 pages of written data and 390 minutes of video (generated by 50 college students aged 18-30 between 2014-2016) this article explores how young people negotiate their own experience and existing discourses about social media. Our analysis shows that young people rely heavily on canonic binaries from utopian and dystopian interpretations of networked technologies to apply labels to themselves, others, and social media in general. As they are prompted to reflect on their experience, their rhetoric about social media use and its implications becomes more nuanced yet remains inherently contradictory. This reflects a dialectical struggle to make sense of their lived experiences and feelings. Our unique methodology for generating deeply self-reflexive, auto-ethnographic narrative accounts suggests a way for scholars to be able to understand the ongoing struggles for meaning that occur within the granularity of everyday reflections about our own social media use.

SCOPUS_ID:85128366732

"it is not your body. It is not your right": Visual arguments of the Argentinean pro-life movement on Instagram

This paper studies the arguments and mobilization strategies used by the pro-life movement in Argentina on Instagram during the debates about the legalization of abortion in 2018. To analyse these strategies, we compiled a corpus, which includes the images published over eleven months by the two accounts of the pro-life associations with most followers on Instagram. The analysis combines two methods: visual content analysis and multimodal critical discourse analysis. Results show that these accounts used different mobilization repertoires (Unidad provida aimed at mobilizing the supporters even at a local level, while Unidos por las dos vidas wanted to persuade the followers, impact them, and stigmatize women). Nevertheless, both accounts converged in the use of the historical argumentation and persuasion strategies of the anti-abortion movement in Argentina, adapting them to the specific affordances of Instagram, without substantially modifying their content. In addition, on occasions, they also constructed visual and textual counterarguments opposed to those of the pro-choice movement.

SCOPUS_ID:85111727928

"it sounds like elves talking"-Polish migrants in Aberystwyth (Wales) and their impressions of the Welsh language

The purpose of this paper was to gain a better understanding of the perceptions of the Welsh language held by the Polish adult migrants in Aberystwyth, Wales. Using qualitative research methods, we collected data from participants concerning their perceptions of the sound and spelling system of Welsh. Data obtained showed that adult Poles in Aberystwyth perceive the phonetics and phonotactics of Welsh to be markedly different from that of their native Polish. The participants believed Welsh to have small number of vowels and large number of consonantal clusters. By comparing consonantal and vowel inventories we were able to demonstrate that Welsh has a more complex vowel inventory than Polish. The consonantal inventories of both languages show great similarities and should not pose major problems to Polish learners of Welsh, who are also speakers of English. As for the phonotactics, Polish possesses a far more complex inventory of consonantal clusters than Welsh. We show that claims of the study's participants that Welsh pronunciation is markedly different from Polish is not based on the linguistic grounds. Instead, such claims must be rooted in the social and ideological perceptions of the Welsh language on the part of the participants in the study.

SCOPUS_ID:85020439135

"it's not comfortable being who i am" - Multilingual identity in superdiverse Dubai

This ethnographic case study examines the factors that contribute to multilingual choices and the construction of identities in a linguistically diverse family within a linguistically diverse city, Dubai in the United Arab Emirates (UAE). Based on interviews with a female Emirati in her early thirties, the article examines this young woman's dispositions (habitus), beliefs and practices with regard to the language and literacy resources at her disposal. It describes how she views English and several varieties of Arabic, focusing largely on her language practices, and the role that they play in the construction and management of her identities, within personal and professional relationships and in relation to the various tasks that she performs in daily life. The study finds that in this superdiverse society, existing language ideologies, indexicalities, stereotypes, and gender issues, may render the performance of multilingualism and the construction of identities potentially rewarding but also potentially disturbing, particularly for those individuals who, through birth and/or marriage, are members of several national or ethnic communities, and who may be, in some senses, icons of superdiversity.

SCOPUS_ID:38349109670

"just don't": The suppression and invitation of dialogue in the mathematics classroom

Responding to concerns raised by grade 11 mathematics students, we examined a broad set of mathematics classroom transcripts from multiple teachers to examine how the word just was and could be used to suppress and invite dialogue. We used corpus linguistics tools to process and quantify the large body of text, not to describe the nature of the discourse, but rather, in the tradition of critical discourse analysis, to prompt reflection on a range of possibilities for directing classroom discourse. We found that the word just was one of the most common words to appear in these classrooms. Drawing on Bakhtin's (The dialogic imagination. Austin: University of Texas Press, 1975/1981) distinctions between monoglossic and heteroglossic utterances, we found that the word just acted as a monoglossic tool, closing down dialogue. We propose, however, that just can also be used as a heteroglossic tool as it can focus attention and thus invite dialogue. © 2007 Springer Science+Business Media B.V.

SCOPUS_ID:85095863481

"keep it Simple, Lazy" - MetaLazy: A New MetaStrategy for Lazy Text Classification

Recent advances in text-related tasks on the Web, such as text (topic) classification and sentiment analysis, have been made possible by exploiting mostly the "rule of more": more data (massive amounts) more computing power, more complex solutions. We propose a shift in the paradigm to do "more with less" by focusing, at maximum extent, just on the task at hand (e.g., classify a single test instance). Accordingly, we propose MetaLazy, a new supervised lazy text classification meta-strategy that greatly extends the scope of lazy solutions. Lazy classifiers postpone the creation of a classification model until a given test instance for decision making is given. MetaLazy exploits new ideas and solutions, which have in common their lazy nature, producing altogether a solution for text classification, which is simpler, more efficient, and less data demanding than new alternatives. It extends and evolves the lazy creation of the model for the test instance by allowing: (i) to dynamically choose the best classifier for the task; (ii) the exploration of distances in the neighborhood of the test document when learning a classification model, thus diminishing the importance of irrelevant training instances; and (iii) a better representational space for training and test documents by augmenting them, in a lazy fashion, with new co-occurrence based features considering just those observed in the specific test instance. In a sizeable experimental evaluation, considering topics and sentiment analysis datasets and nine baselines, we show that our MetaLazy instantiations are among the top performers in most situations, even when compared to state-of-the-art deep learning classifiers such as Deep Network Transformer Architectures.

SCOPUS_ID:85050586366

"language" and "discourse": Two perspectives on linguistic philosophy

With the establishment of modern linguistics and the linguistic turn of western philosophy, various linguistic theories have been advanced and have given different interpretations to language and discourse. Different schools of thought have witnessed a direct collision of ideas and a deep academic dialogue between the theory of translinguistics advanced by the great master of dialogism, Bakhtin, and the outlook on language of the father of modern linguistics, Saussure.

SCOPUS_ID:85124239386

"nature needs you": Discursive constructions of legitimacy and identification in environmental charity appeals

This study traces how discursive constructions of legitimacy and identification are enacted textually and visually with respect to environment-oriented causes, such as landscape or species restoration. Such conservation projects actually clash with human economic priorities typical of the Anthropocene. Drawing on models of social trust and assuming the discursive nature of legitimacy and identification, we explore how environmental charity organizations represent their conservation efforts, reproduce sustainability discourses and advocate self-regulatory practices. We use a sample of mission statements and donation appeals by six prominent environmental charities from the UK. Through keyness and concordance analysis, we identify textual strategies that position the prospective donor as a "beneficiary"of environment-oriented actions. We also analyze rhetorical strategies and visual resources that align the aims of the organization with the social imaginaries and emotional dispositions of prospective donors.

SCOPUS_ID:85091304219

"nobody Speaks that Fast!" An Empirical Study of Speech Rate in Conversational Agents for People with Vision Impairments

The number of people with vision impairments using Conversational Agents (CAs) has increased because of the potential of this technology to support them. As many visually impaired people are accustomed to understanding fast speech, most screen readers or voice assistant systems offer speech rate settings. However, current CAs are designed to interact at a human-like speech rate without considering their accessibility. In this study, we tried to understand how people with vision impairments use CA at a fast speech rate. We conducted a 20-day in-home study that examined the CA use of 10 visually impaired people at default and fast speech rates. We investigated the difference in visually impaired people's CA use with different speech rates and their perception toward CA at each rate. Based on these findings, we suggest considerations for the future design of CA speech rate for those with visual impairments.

SCOPUS_ID:85117954651

"pocketBot Is like a Knock-On-the-Door!": Designing a Chatbot to Support Long-Distance Relationships

Many couples experience long-distance relationships (LDRs), and "couple technologies"have been designed to influence certain relational practices or maintain them in challenging situations. Chatbots show great potential in mediating people's interactions. However, little is known about whether and how chatbots can be desirable and effective for mediating LDRs. In this paper, we conducted a two-phase study to design and evaluate a chatbot, PocketBot, that aims to provide effective interventions for LDRs. In Phase I, we adopted an iterative design process through conducting need-finding interviews to formulate design ideas and piloted the implemented PocketBot with 11 participants. In Phase II, we evaluated PocketBot with eighteen participants (nine LDR couples)in a week-long field trial followed by exit interviews, which yielded empirical understandings of the feasibility, effectiveness, and potential pitfalls of using PocketBot. First, a knock-on-the-door feature allowed couples to know when to resume an interaction after evading a conflict; this feature was preferred by certain participants (e.g., participants with stoic personalities). Second, a humor feature was introduced to spice up couples' conversations. This feature was favored by all participants, although some couples' perceptions of the feature varied due to their different cultural or language backgrounds. Third, a deep talk feature enabled couples at different relational stages to conduct opportunistic conversations about sensitive topics for exploring unknowns about each other, which resulted in surprising discoveries between couples who have been in relationships for years. Our findings provide inspiration for future conversational-based couple technologies that support emotional communication.

SCOPUS_ID:84948175759

"potentialities or possibilities": Towards quantum information science?

The use of quantum concepts and formalisms in the information sciences is assessed through an analysis of published literature. Five categories are identified: use of loose analogies and metaphors between concepts in quantum physics and library/information science; use of quantum concepts and formalisms in information retrieval; use of quantum concepts and formalisms in studying meaning and concepts; quantum social science, in areas adjacent to information science; and the qualitative application of quantum concepts in the information disciplines. Quantum issues have led to demonstrable progress in information retrieval and semantic modelling, with less clear-cut progress elsewhere. Whether there may be a future "quantum turn" in the information sciences is debated, the implications of such a turn are considered, and a research agenda outlined.

SCOPUS_ID:85030706232

"reservoir of rage swamps Wall St" the linguistic construction and evaluation of Occupy in international print media

Originating on New York's Wall Street, the Occupy movement was "an international network of protests against social and economic inequality that began in [September] 2011 in response to the downturn of 2008" (Thorson et al. 2013, 427). Whilst there has been research on online activity in relation to Occupy, the scope of linguistic analysis to date has been somewhat narrow. Furthermore, the focus on new media has indirectly led to an absence of analysis of institutionally-endorsed traditional media texts. We adopt a mixed-method approach of corpus analysis and discourse analysis of national newspaper articles to answer questions such as 'Is Occupy associated with a semantic field of violence and aggression?' and 'Who is represented as having agency?' Our results indicate that, in our small corpus of media texts, Occupy and its supporters were predominantly portrayed negatively at the movement's height; even though protesters are reported to have been peaceful in their majority, the English-speaking media we analysed still aligns them with language suggestive of aggression, conflict and even violence.

SCOPUS_ID:85020040337

"same News, Different Stances"? A comparative media discourse investigation of hard news texts in the new straits times and berita harian

This paper investigates news media texts in the Malay- and English-language print media in Malaysia. We analyse 'hard news' reports covering the same story in Malay and English from the New Straits Times (NST) and Berita Harian (BH). Kaplan's early studies on contrastive rhetoric (1966, 1987, 1988) suggest that cross-language differences in paragraph organisation may reflect differences in thinking or at least differences in writing conventions that are learnt in a culture. Thus, this study hopes to investigate to what extent this applies to Malay and English media texts. Using a modified CDA framework, a 'product' approach is applied in order to establish the degree of parallelism between the Malay and English media texts reporting the same story, and the degree of translation equivalence. A 'process' approach based on interviews is also used in order to discover the policies and processes involved in the construction of print media texts in both languages. The findings reveal that although there are commonalities in terms of structure and stance between the hard news texts found in both papers, there is some evidence of different stances adopted by the editors and journalists of the NST and the BH in terms of their inclusion of detail and their level of involvement or detachment in reporting crime and accident stories.

SCOPUS_ID:85084388763

"so you know ehn.. " the use of bilingual interjections in Nigerian English

This paper investigates four bilingual interjections: na wa, shikena, ehn, and ehen, with the objective of exploring their sources, meanings, frequencies, spelling stability, positions, collocational patterns and discourse-pragmatic functions in Nigerian English. The data which were obtained from the Nigerian component of the Global Web-based English corpus were analyzed quantitatively and qualitatively, using the theory of pragmatic borrowing. The results indicate that na wa, which is loaned from Nigerian Pidgin, is actually a modified form of a Hausa expression, na wahala, shikena is borrowed from Hausa, while ehn and ehen are loaned from Yoruba. Na wa is an emotive interjection, shikena and ehen are cognitive interjections, while ehn can function both as phatic and as emotive interjections. Both ehn and ehen also function as pragmatic markers. The study thus extends research on the discourse-pragmatic features of Nigerian English.

SCOPUS_ID:77952438527

"the greatest man whom Germany had in recent times"- Herder's relationship with Leibniz

In May 1785 Georg Forster wrote to a friend: "Herder is quite a Leibnizian thinker". The following paper on Herder's reception of Leibniz I have read in the Leibniz Society on April 1, 2004. Previously to that time Herder scholars traditionally had prefered Spinoza as the favourite philosopher of Herder's. As for Leibniz they had in view only Herder's natural philosophy. - First I describe some documents of reception: manuscripts, sermons and letters, especially of the young Herder. After that the fundamental conceptions and principal thoughts of Herder's chief works are reduced to Leibnizian ideas, concerning the theory of language and cognition, the individuality of man and the immortality of the soul, the unity of natural and human history, the organic forces, evolution and the "laws of nature". Finally Herder's late essays on Leibniz are compared with their literary sources which have been unknown hitherto. ©Franz Steiner Verlag, Stuttgart.

SCOPUS_ID:85042848620

"the oriental moon over the coast of the west" by Moudarres and Khaznadar as a visual text

This paper deals with the artistic structure of Moudarres & Khaznadar's collection entitled The Oriental Lune over the Coast of the West 'the Greek passage' published in Damascus in 1962. In which words and painting met in poems written both in Arabic and French. The Artistic shape of the collection was affected by the vision of a poet who is a fine artist. This effect gave the visual shape a distinction in the poetic collections of the Arab world. The paintings played a major role in enriching the content of the words and made the poems visual texts which cannot be received properly by merely listening to them. So reading is the suitable channel of reception, especially since the authors use visual effects such as enlargement, minimizing and handwriting in addition to paintings.

SCOPUS_ID:84994692102

"the semantics of migration". Translation as Transduction: Remaking meanings across modes

This paper adopts a multimodal critical discourse analysis (MCDA) approach to analyse how meanings are produced and circulated in British major corporate digital media outlets via the multimodal notion of transduction (Kress 1997; Mavers 2011; Newfi eld 2014). Transduction is a form of translation from one semiotic system to another one, for example from verbal language to images and vice versa. However, transductions cannot be interpreted as mere transferrals from one resource to another one, and are here interpreted as multiplying meanings (Lemke 2002). As a case study, this paper will select some online columns from the Telegraph and the Guardian, drawing from a monitor corpus that is under construction to date and that includes multimodal data from the British digital press reporting on the "European migrant crisis" in 2015. The columns selected for this study deal with how people on the move are and/ or should be labelled (e.g. Migrants? Refugees? Asylum seekers? Potential terrorists? See Gabrielatos, Baker 2008; Baker et al. 2008). The columns will be commented qualitatively from a multimodal critical discourse framework of analysis, with the goal of shedding light on how pictorial materials (e.g. pictures and diagrams) can amplify, reduce or even contradict what is argued in the verbal text. In the conclusive remarks, some refl ections will be presented with a view to possible future lines of research.

SCOPUS_ID:85108773734

"there are no words that are 'clear' in and of themselves": Meta-pragmatic comments and semantic analysis in legal interpretation

Legal interpretation often includes a profusion of meta-pragmatic comments about the interpretation process itself. Thus, while pragmatic theories refer to the interpretation processes as natural, mostly unconscious processes, in legal interpretation the inference processes take on a conscious form. Meta-pragmatic comments provide a glimpse into this process and surface various aspects of it that have been described theoretically. The aim of this study is to examine the possibility of applying theoretical pragmatic terms to the legal interpretation discourse.A semantic-pragmatic analysis of a few cases shows that while the linguistic component of the legal interpretation makes it easy to apply pragmatic theory, some of the procedures performed by judges are incompatible with a semantic-pragmatic interpretation and contradict its theoretical assumptions. The purposive approach to interpretation that has developed in the Israeli legal system raises some serious problems in that sense. Applying the objective purpose of a statute even when it is obvious that the legislators could not have desired that in order to change the law cannot be considered "interpretation"in the pragmatic sense, since the central element of speaker's intentions has been given a completely different meaning, and the aim of the interpretation procedure, namely identifying the speaker's intentions, has for the most part been lost. This paper suggests that these cases should be viewed as exceptions to pragmatic interpretation and they do not permit application of pragmatic theory to them, at least not of the types of approaches attributed to Grice's legacy. Nevertheless, this does not mean that theoretical pragmatic tools are not applicable to judicial opinions and to other kinds of legal text.

SCOPUS_ID:38049051876

"things get glossed over": Rearticulating the silencing power of whiteness in education

This article investigates the ways that White teachers approach issues of race, racism, and White supremacy in White-dominated educational settings. Drawing from data from a yearlong qualitative research study, the article uses discourse analysis, critical studies of Whiteness, and feminist theory to detail 15 rhetorical, behavioral, analytical, and interactional strategies that participants used to insulate themselves from implication in social inequality. The article demonstrates how participation in these strategies stymied attempts at transformative multicultural education and thus functioned to reproduce, rather than challenge, the status quo of educational and social inequality. © 2008 by the American Association of Colleges for Teacher Education.

SCOPUS_ID:84955589853

"twitter archeology" of learning analytics and knowledge conferences

The goal of the present study was to uncover new insights about the learning analytics community by analyzing Twitter archives from the past four Learning Analytics and Knowledge (LAK) conferences. Through descriptive analysis, in- teraction network analysis, hashtag analysis, and topic modeling, we found: extended coverage of the community over the years; increasing interactions among its members regard- less of peripheral and in-persistent participation; increasingly dense, connected and balanced social networks; and more and more diverse research topics. Detailed inspection of semantic topics uncovered insights complementary to the analysis of LAK publications in previous research.

SCOPUS_ID:85073063337

"units of Meaning" in Medical Documents: A Natural Language Processing Perspective

This paper discusses principles for the design of natural language processing (NLP) systems to automatically extract data from doctor's notes, laboratory results and other medical documents in free-form text. We argue that rather than searching for "atom units of meaning" in the text and then trying to generalize them into a broader set of documents through increasingly complicated system of rules, an NLP practitioner should take concepts as a whole and as a meaningful unit of text. This simplifies the rules and makes NLP system easier to maintain and adapt. The departure point is purely practical; however, a deeper investigation of typical problems with the implementation of such systems leads us to a discussion of broader linguistic theories underlying the NLP practices, such as metaphors theories and models of human communication.

SCOPUS_ID:85020398442

"verbal and nonverbal" in semiotics

The "verbal-nonverbal" distinction is mostly used in everyday language and its "'naïve-natural' attitude" (Husserl). It confirms the idea that a word/verb, as a component of human expressivity, is the basic unit of language. Theories of Peirce, Saumjan, and Searle highlight how a different, predominantly "'non-naïve'-natural attitude" is required to understand the distinction and its position in the semiotic toolkit. To support this conclusion, Husserl unfolds a methodological approach of varying attitudes and attitude-changes, including important diversifications of ontology. A consequence is the need for an interregional ontological approach, which in this article leads to a consideration of social psychology (Lewin) and quantum theory (Bohm) because both underline that words and meanings are forces in fields, and by no means isolated single units. Word and meaning are to be understood as forces, and meaning-making as well as interpretation a matter of force field considerations. Semiotics should thus cherish dynamic features, whereby the "verbal-nonverbal" distinction teaches us at a "non-naïve" attitude level, that a word/verb is always a non-word/verb as well. The greatness of semiotics is in the understanding of such dynamic and continuously creative inversions.

SCOPUS_ID:85130368498

"we just want the language tone": When requests to use minority languages lead to interactional breakdown in multilingual classrooms

In this paper, we analyze two instances of interactional breakdown in linguistically and culturally diverse classrooms in Copenhagen and Helsinki. Our focus is situations where teachers request the use of minority languages from pupils, and pupils react reluctantly and display embarrassment. These situations represent sociolinguistic spaces of upset understood as disruptions of prevailing language ideologies and sociolinguistic regimes. We argue that pupils' reluctance to comply with teachers' attempts to include minority languages exemplifies such a disruption, and meta-communicative exchanges represent a window into the language ideologies influencing such situations. We analyze the interactions sequentially through the theoretical lens of enregisterment, linguistic legitimacy, and raciolinguistic microaggressions. The Helsinki data are drawn from a sociolinguistically informed action research project in an elementary school. The Copenhagen data involve lower-secondary-level pupils and consist of observations and recordings collected as part of a long-term ethnographic study. Despite the differences in projects and field sites, we found a striking similarity both in the language ideologies displayed by teachers and in pupils' reactional patterns. Consequently, we argue that both examples represent the same type of sociolinguistic space of upset characterized by an intrinsic dilemma in Nordic public schools, which are simultaneously expected to secure the continuation of mainstream culture and embrace linguistic diversity.

SCOPUS_ID:84971642445

"we should⋯" versus "we will⋯": How do the governments report their work in "one Country Two Systems"? A corpus-driven critical discourse analysis of government work reports in Greater China

Critical discourse analysis (CDA) in conjunction with a corpus-driven analytical methodology has evolved into a powerful qualitative and quantitative tool for deconstructing and studying political discourse. This study utilizes a corpus-driven CDA approach to examine the dynamics of power distance and the ideological stance in the context of Greater China, as conveyed in the 2013 Report on the Work of the Government of Mainland China and the Policy Address in Hong Kong. Concordancing software was used to generate frequency lists, co-selection patterns, and concgrams for detailed analysis. In particular, study examines differences in usage of the first-person plural pronoun we collocated with modal verbs and related lexical items. Concgrams in discourse analysis offer insights into the discursive practice of political actors in this unique political discourse genre. The findings show that the distribution and utilization of first-person plural pronoun we, and its interplay with other modal verbs and lexical items, comprising a specific "concgram," have provided textual and intertextual evidence for the analytical results of conventional critical discourse analysis. The discussions support that a methodological synergy between corpus linguistics and critical discourse analysis can serve as a powerful tool to deconstruct and analyze political discourses.

SCOPUS_ID:85109017198

"we thought about it together and the solution came to our minds": Languaging linguistic problem-solving in multilingual Finnish classrooms

This study examines a learning experiment in which linguistic problem-solving tasks designed to increase students' (aged 9-13) language awareness through collaborative dialogue were introduced in multilingual primary school classrooms in Finland. The aim was to analyse how the students (N = 126) reported what was happening during the linguistic problem-solving tasks, drawing on the method of languaging. Additionally, the study investigates how meaningful, relevant and novel the students with diverse backgrounds found the tasks. The data were collected via a survey. Students' problem-solving reports were analysed via content analysis, with the Taxonomy of Cognitive Process applied. Statistical analysis was used to measure the experienced meaningfulness, relevance and novelty. The analysis resulted in an understanding of the multiple voices in which the students articulated their thinking regarding linguistic problem-solving. The study sheds light on how to develop language aware learning materials to engage all students, regardless of their backgrounds, in discussions on language.

SCOPUS_ID:85121420037

"what i want to do i do not do": On bi: On multilingual repertoires and linguistic dislocation in a border town

Language problems and language barriers are challenges facing not only immigrants but also minorities and people in rural/semirural areas. This study examines individuals' bi- and multilingual repertoires, language practices and attitudes in a Hokkien-speaking community in Kangar, a semirural town of northern Malaysia bordering Thailand. Through questionnaire surveys and interviews, we investigate how these notions can be used as a means to understand/reflect bilingualism and multilingualism and, more importantly, the potential disparity between what people want to do/say and what people eventually manage to do/say. While there is a shift in language practice from a local- and ancestral origin-induced pattern towards a more "global"and "pan-Chinese"paradigm, the findings also reveal the linguistic "dislocations"of the Hokkien-speaking community across ALL generations regardless of ethnicity. The language issues in the community reflect - and are likely to be reflections of - society at large. The vast contrast between individual/societal linguistic aspirations and the actual linguistic repertoire/communicative competence among the locals indicates the need to redress an absence of major efforts to close urban-rural/city-town/dominant-dominated social divides across the (language) education landscape at the national level.

SCOPUS_ID:85088836539

"why Don't You Act like You Believe It?": Competing Visions of Climate Hypocrisy

This paper interrogates how the notion of hypocrisy is invoked in relation to climate change and offers two key findings. First, it demonstrates that invocations of hypocrisy are not only deployed by conservative opponents of climate action, but also by progressive proponents of such action. Second, this article shows that while hypocrisy discourse is used to support both anti- and pro-climate change perspectives, its nature and function fundamentally differs depending on who is using it. The article identifies four discrete types of climate hypocrisy discourse. Conservatives who reject climate change action tend to use two "modes"of hypocrisy discourse. The first is an "individual lifestyle outrage"mode that cultivates outrage about the hypocritical behavior and lifestyle choices of climate activists to undermine the urgency and moral need for climate change action. The second, an "institutional cynicism"mode, encourages a cynical fatalism about any proposed governmental action regarding climate change by suggesting that governments are necessarily climate hypocrites because of the economic and political impossibility of serious emissions reductions. In contrast, progressives use hypocrisy discourse in two different modes. The first involve an "institutional call to action"mode that uses charges of hypocrisy to attack government inaction on climate change and demand that effective action be taken in line with their public commitment to climate action. Secondly, they also employ a "reflexive"mode in which explorations of the ubiquity of climate change hypocrisy illuminate the dilemmas that virtually all responses to climate change necessarily grapple with in our current context. Overall, the article seeks to contribute to our understanding of climate change communications by (i) showing that hypocrisy discourse is not simply a sensationalist PR strategy of conservatives but is rather a broad, significant and multi-faceted form of climate change discourse; and (ii) suggesting that certain modes of hypocrisy discourse might not only represent genuine attempts to make sense of some of the fundamental tensions of climate change politics but also help us understand the challenge that the "entanglement"of personal agency/choice within broader political structures presents, and thus heighten positive affective commitments to climate change action.

SCOPUS_ID:85045071953

"you are struggling forwards, and you don't know, and then you ⋯ you do code-switching⋯" - Code-switching in ELF Skype conversations

This article analyzes how code-switching (CS) is used as a key strategy in English as a Lingua Franca (ELF) interaction. We use data from CASE, a corpus of ELF Skype conversations in an informal setting between students from nine European countries. CS in our data is commonly used as a communication strategy. A quantitative analysis shows that it occurs in the majority of conversations and across all nationalities. Participants mostly switch to their own L1, frequently also to their interlocutors' L1, rarely to other languages. CS co-occurs with other discourse features, in particular other instances of CS and laughter. Based on a qualitative analysis, we distinguish three motivations and functions of CS: On an interpersonal level, CS can underline group membership or cultural identity and create rapport, often in combination with laughter. CS is used to improve communication at the discourse level by conveying concepts that are untranslatable and possibly unknown to the interlocutor, in metalinguistic commentary, or to close lexical gaps. CS can also be used for addressee specification. Through its combined qualitative and quantitative approach, the article aims at contributing to the analysis of the context as well as the motivations and functions of CS in ELF.

SCOPUS_ID:84929080649

"¡Chávez vive...!": La sacralización del líder como estrategia en el discurso político venezolano

The sacralization of politics is a strategy of institutional political discourse oriented towards the continuity and legitimation of governments, and used by regimes which prescribe a unique line of thought and deny the autonomy of the individual versus the collectivity (Gentile 1990, 1993). This article studies the sacralization of the leader as observed in Venezuela during the government of Hugo Chávez and after his death. Critical Discourse Analysis was applied to texts both of the president and of his successors, in order to show how an incipient political religion is constructed in Venezuela based on the Bolivarian revolution and its eternal commander. We suggest that this tendency has its origin in the siege of power, and is so distinct from popular creeds.

SCOPUS_ID:85012922102

"¡Qué amables tus moradas..." (Psalm 84 [Vg 83]), by fray Pedro Malón de Echaide, and fray Luis de León's model for biblical psalms translation

In this paper I propose an analysis of the biblical paraphrases included in La conversión de la Madalena (Barcelona, Hubert Gotard, 1588), by fray Pedro Malón de Echaide. In this ascetic treatise the Augustinian writer inserts several poems, most of them translations or, rather, exegetical paraphrases of different biblical Psalms, following the model of fray Luis de León. My commentary focuses on the poem that begins "¡Qué amables tus moradas...", placed in the "Parte primera del tratado de la Madalena", § 1, a parafrase of Psalm 84 (Vulgata 83), "Quam dilecta tabernacula tua, Domine virtutum". I offer an annotated version of the text and analyse its content and structure, as well as the amplificatio technique used by Malón de Echaide in his translation and the main rhetorical devices.

SCOPUS_ID:84991829681

"¡Qué amables tus moradas⋯" (Psalm 84 [Vg 83]), by fray Pedro Malón de Echaide, and fray Luis de León's model for biblical psalms translation

In this paper I propose an analysis of the biblical paraphrases included in La conversión de la Madalena (Barcelona, Hubert Gotard, 1588), by fray Pedro Malón de Echaide. In this ascetic treatise the Augustinian writer inserts several poems, most of them translations or, rather, exegetical paraphrases of different biblical Psalms, following the model of fray Luis de León. My commentary focuses on the poem that begins "¡Qué amables tus moradas⋯", placed in the "Parte primera del tratado de la Madalena", § 1, a parafrase of Psalm 84 (Vulgata 83), "Quam dilecta tabernacula tua, Domine virtutum". I offer an annotated version of the text and analyse its content and structure, as well as the amplificatio technique used by Malón de Echaide in his translation and the main rhetorical devices.

SCOPUS_ID:84901926685

"¿Cómo estas?" "i'm good." Conversational code-switching is related to profiles of expressive and receptive proficiency in Spanish-English bilingual toddlers

Relations between bilingual children's patterns of conversational code-switching (responding to one language with another), the balance of their dual language input, and their expressive and receptive proficiency in two languages were examined in 115 2-year-old simultaneous Spanish-English bilinguals in the U.S. Children were more likely to code-switch in response to Spanish than English. Children's expressive vocabulary scores were higher in English than in Spanish, while their English and Spanish receptive language scores were not different. Analyses of subgroups of children with different but consistent patterns of code-switching confirmed that children who code-switched to English showed greater English skills, specifically in the expressive domain. Children who did not code-switch were more balanced bilinguals in both expressive and receptive skills. Children with other code-switching patterns showed still different profiles of dual language expressive and receptive proficiency. These findings reveal that some, but not all, bilingual children show different profiles of expressive and receptive skill in their two languages and that these proficiency profiles are related to their language choices in conversation. © The Author(s) 2014.

SCOPUS_ID:85068107709

#BiggBoss—Long-Run Event Detection and Sentiment Mining in Twitter

Online social media like Twitter and Facebook provides a common platform for presenting views and opinions of an individual on various events. This research work aims at detecting peak events that occur in online social networks using the proposed approach of exponential moving average algorithm and peak recognition method. A two-level hybrid sentiment analysis using n-grams and Naïve Bayes classifiers is performed on tweets to ensure the true sentiment of the user. The analysis was based on bag-of-words model and Bill McDonald’s list for positive and negative words. The tweets were streamed for #BiggBoss and stored in sequential bins. The potential event was detected on the final day of result announcement by peak recognition method. The Naïve Bayes classifier predicted tweets with accuracy of 89% which would further aid in event summarization and eliminate event-related rumors.

http://arxiv.org/abs/2007.14936v1

#Brexit: Leave or Remain? The Role of User's Community and Diachronic Evolution on Stance Detection

Interest has grown around the classification of stance that users assume within online debates in recent years. Stance has been usually addressed by considering users posts in isolation, while social studies highlight that social communities may contribute to influence users' opinion. Furthermore, stance should be studied in a diachronic perspective, since it could help to shed light on users' opinion shift dynamics that can be recorded during the debate. We analyzed the political discussion in UK about the BREXIT referendum on Twitter, proposing a novel approach and annotation schema for stance detection, with the main aim of investigating the role of features related to social network community and diachronic stance evolution. Classification experiments show that such features provide very useful clues for detecting stance.

SCOPUS_ID:85028976222

#Communing affiliation: Social tagging as a resource for aligning around values in social media

Social metadata is an important dimension of social media communication, and closely associated with practices such as curating, tagging, and searching content. This article explores how hashtags are used to coordinate and accentuate the values construed in a corpus of Twitter posts (tweets) about depression. In other words, it explores how people use hashtags as a resource to ‘convoke’ communities of feeling around values realised as ideation-attitude couplings. The aim is to extend work on dialogic affiliation (Knight, 2010a, 2010b, 2013) in order to account for ‘ambient’ affiliation via social media, where participants do not necessarily interact directly. We employ a discursive system, communing affiliation, to interpret how particular values about depression are positioned as bondable in this ambient environment. The focus is on understanding how people are forging alignments and negotiating meaning through social tagging practices.

SCOPUS_ID:85082400257

#Confused and beyond: Detecting confusion in course forums using students' hashtags

Students' confusion is a barrier for learning, contributing to loss of motivation and to disengagement with course materials. However, detecting students' confusion in large-scale courses is both time and resource intensive. This paper provides a new approach for confusion detection in online forums that is based on harnessing the power of students' self-reported affective states (reported using a set of pre-defined hashtags). It presents a rule for labeling confusion, based on students' hashtags in their posts, that is shown to align with teachers' judgement. We use this labeling rule to inform the design of an automated classifier for confusion detection for the case when there are no self-reported hashtags present in the test set. We demonstrate this approach in a large scale Biology course using the Nota Bene annotation platform. This work lays the foundation to empower teachers with better support tools for detecting and alleviating confusion in online courses.

http://arxiv.org/abs/2005.08224v1

#Coronavirus or #Chinesevirus?!: Understanding the negative sentiment reflected in Tweets with racist hashtags across the development of COVID-19

Situated in the global outbreak of COVID-19, our study enriches the discussion concerning the emergent racism and xenophobia on social media. With big data extracted from Twitter, we focus on the analysis of negative sentiment reflected in tweets marked with racist hashtags, as racism and xenophobia are more likely to be delivered via the negative sentiment. Especially, we propose a stage-based approach to capture how the negative sentiment changes along with the three development stages of COVID-19, under which it transformed from a domestic epidemic into an international public health emergency and later, into the global pandemic. At each stage, sentiment analysis enables us to recognize the negative sentiment from tweets with racist hashtags, and keyword extraction allows for the discovery of themes in the expression of negative sentiment by these tweets. Under this public health crisis of human beings, this stage-based approach enables us to provide policy suggestions for the enactment of stage-specific intervention strategies to combat racism and xenophobia on social media in a more effective way.

SCOPUS_ID:85124605216

#CoronavirusCruise: Impact and implications of the COVID-19 outbreaks on the perception of cruise tourism

Early in the COVID-19 pandemic, the Diamond Princess became the center of the largest outbreak outside the original epicenter in China. This outbreak which left 712 passengers infected and 14 dead, followed by subsequent outbreaks affecting over one-third of the active ships in the cruise industry's global fleet, quickly became a crisis that captured public attention and dominated mainstream news and social media. This study investigates the perception of cruising during these outbreaks by analyzing the tweets on cruising using Natural Language Processing (NLP). The findings show a prevalent negative sentiment in most of the analyzed tweets, while the criticisms directed at the cruise industry were based on perceptions and stereotypes of the industry before the pandemic. The study provides insight into the concerns raised in these conversations and highlights the need for new business models outside the pre-pandemic mass-market model and to genuinely make cruising more environmentally friendly.

SCOPUS_ID:84987745734

#Criming and #Alive: Network and content analysis of two sides of a story on twitter

On December 3, 2014, after a grand jury decided not to indict the white police officer in the death of Eric Garner, the social networking platform Twitter was flooded with tweets sharing stances on racial profiling and police brutality. To examine how issues concerning race were communicated and exchanged during this time, this study compares differences between tweets using two trending hashtags #CrimingWhileWhite (#cww) and #AliveWhileBlack (#awb) from December 3 through December 11, 2014. To this end, network and content analysis are used on a large dataset of tweets containing the hashtags #awb and #cww. Findings indicate that there are clear differences, both structurally and in linguistic style, between how individuals express themselves based on which hashtag they used. Specifically, we found that #cww users disproportionately shared informational content, which may have led to the hashtag gaining more network volume and attention as a trending topic than #awb. In contrast, #awb tweets tended to be more subjective, expressing a sense of community and strong negative sentiment toward persistent structural racism.

SCOPUS_ID:85114433999

#CybillToo?: how a feminist sitcom (almost) exposed Hollywood’s dark secrets

The sitcom Cybill (CBS 1995–1998) is loosely based on the real-life experiences of its feminist star and executive producer, Cybill Shepherd. The show thus has the extraordinary potential to relate the experiences of a Hollywood actress, operating within a singularly influential apparatus in the creation of dominant ideologies, including representations of women. This ‘belly of the beast’ set-up suggests an abundance of possibilities for the woman buffoon, trickster or thief of language to tell her thus-far censored, inside story, but within a medium and textual format that are themselves significantly controlled by, and instrumental in perpetuating, that very same machinery. Yet this sitcom was produced in the nineteen-nineties, a decade characterised by apolitical, postmodern irreverence. This article will assess the sitcom’s integration of postmodern, conservative and feminist discourses by conducting a critical discourse analysis, as formulated by Norman Fairclough, of one episode. The episode is one of two in which the show depicts Hollywood producers requesting sexual favours in return for employment opportunities, two decades before the practice provoked international outrage and brought about the formation of the #MeToo movement. The episode overall adopts a morally dubious positioning in its portrayal of Hollywood realities. However, as will be argued, this very postmodern ambiguity enabled the sitcom to represent and denounce long-established ‘casting couch’ practices, if in a cursory and trivialising manner.

SCOPUS_ID:84951954478

#DigitalHealth: Exploring users' perspectives through social media analysis

In order to explore the role of social media in forming an understanding of digital healthcare, we conducted a study involving sentiment and network analysis of Twitter contents. In doing this, we gathered 20,400 tweets that mentioned the key term #DigitalHealth for 55 hours, over a three-day period. In addition to examining users' opinions through sentiment analysis, we calculated in-degree centralities of nodes to identify the hubs in the network of interactions. The results suggest that the overall opinion about digital healthcare is generally positive. Additionally, our findings indicate that the most prevalent keywords, associated with digital health, widely range from mobile health to wearable technologies and big data. Surprisingly, the results show that the newly announced wearable technologies could occupy the majority of discussions.

SCOPUS_ID:85034268419

#ForgiveUsForWeHaveSinned: Conceptual integration theory and political Internet humour

The aim of the paper is to uncover the extent to which different forms of political Internet humour can criticise current political affairs in a developing democracy such as Bosnia and Herzegovina. Specifically, applying a cognitive linguistic theory of meaning construction, namely conceptual integration theory, the paper analyses the construction of meaning of humorous Internet forms, such as memes, demotivational posters, hashtag posts, and memetic photographs, representing innovative ways of providing political commentaries on current political affairs. The meaning of political humour is constructed in conceptual blending as a basic cognitive mechanism. As it is claimed (Coulson & Pascual 2006, Coulson & Oakley 2006, Coulson 2006, Oakley & Coulson 2008) that blending can be used as a rhetorical tool influencing the audience to change the reality and even act upon it, the analysis of the construction of meaning of political humour as products of conceptual integration can reveal hidden ideologies in political discourse.

SCOPUS_ID:85046744997

#GirlsLikeUs: Trans advocacy and community building online

In this research, we examine the advocacy and community building of transgender women on Twitter through methods of network and discourse analysis and the theory of networked counterpublics. By highlighting the network structure and discursive meaning making of the #GirlsLikeUs network, we argue that the digital labor of trans women, especially trans women of color, represents the vanguard of struggles over self-definition. We find that trans women on Twitter, led by Janet Mock and Laverne Cox, and in response to histories of misrepresentation and ongoing marginalization and violence, deliberately curate an intersectional networked counterpublic that works to legitimize and support trans identities and advocate for trans autonomy in larger publics and counterpublics.

SCOPUS_ID:85069849591

#Globalcitizen: An explorative Twitter analysis of global identity and sustainability communication

(1) Background: global citizenship is often associated with pro-social and pro-environmental attitudes, beliefs and behaviors. Much of this research relies on questionnaire studies, whereas studies drawing on naturally occurring daily communications are under-used. In this paper, we analyse the content that users publish on Twitter related to the issue of global identity and citizenship. (2) Methods: we assessed word frequencies and associated hashtags of 35,237 tweets marked with the hashtag #globalcitizen. A sentiment analysis was conducted to investigate the moods and emotions of the tweets. (3) Results: in line with expectations derived from social identity theory, we found that associated words and hashtags were more often linked to themes of common good/disadvantaged groups than they were to the topic of nature and environment. Providing evidence for an empowerment function of global citizenship, the sentiment analysis suggests that global citizenship is related to rather positive emotions. (4) Conclusion: these findings reveal how identity constructions in social media predict associated contents and possibly pathways to social change.

SCOPUS_ID:85066241888

#Healthy selfies: Exploration of health topics on instagram

Background: Social media provides a complementary source of information for public health surveillance. The dominate data source for this type of monitoring is the microblogging platform Twitter, which is convenient due to the free availability of public data. Less is known about the utility of other social media platforms, despite their popularity. Objective: This work aims to characterize the health topics that are prominently discussed in the image-sharing platform Instagram, as a step toward understanding how this data might be used for public health research. Methods: The study uses a topic modeling approach to discover topics in a dataset of 96,426 Instagram posts containing hashtags related to health. We use a polylingual topic model, initially developed for datasets in different natural languages, to model different modalities of data: hashtags, caption words, and image tags automatically extracted using a computer vision tool. Results: We identified 47 health-related topics in the data (kappa=.77), covering ten broad categories: acute illness, alternative medicine, chronic illness and pain, diet, exercise, health care & medicine, mental health, musculoskeletal health and dermatology, sleep, and substance use. The most prevalent topics were related to diet (8,293/96,426; 8.6% of posts) and exercise (7,328/96,426; 7.6% of posts). Conclusions: A large and diverse set of health topics are discussed in Instagram. The extracted image tags were generally too coarse and noisy to be used for identifying posts but were in some cases accurate for identifying images relevant to studying diet and substance use. Instagram shows potential as a source of public health information, though limitations in data collection and metadata availability may limit its use in comparison to platforms like Twitter.

SCOPUS_ID:85141692714

#IStandWithPutin Versus #IStandWithUkraine: The Interaction of Bots and Humans in Discussion of the Russia/Ukraine War

The 2022 Russian invasion of Ukraine emphasises the role social media plays in modern-day warfare, with conflict occurring in both the physical and information environments. There is a large body of work on identifying malicious cyber-activity, but less focusing on the effect this activity has on the overall conversation, especially with regards to the Russia/Ukraine Conflict. Here, we employ a variety of techniques including information theoretic measures, sentiment and linguistic analysis, and time series techniques to understand how bot activity influences wider online discourse. By aggregating account groups we find significant information flows from bot-like accounts to non-bot accounts with behaviour differing between sides. Pro-Russian non-bot accounts are most influential overall, with information flows to a variety of other account groups. No significant outward flows exist from pro-Ukrainian non-bot accounts, with significant flows from pro-Ukrainian bot accounts into pro-Ukrainian non-bot accounts. We find that bot activity drives an increase in conversations surrounding angst (with p= 2.450 × 10- 4 ) as well as those surrounding work/governance (with p= 3.803 × 10- 18 ). Bot activity also shows a significant relationship with non-bot sentiment (with p= 3.760 × 10- 4 ), where we find the relationship holds in both directions. This work extends and combines existing techniques to quantify how bots are influencing people in the online conversation around the Russia/Ukraine invasion. It opens up avenues for researchers to understand quantitatively how these malicious campaigns operate, and what makes them impactful.

SCOPUS_ID:85045062619

#London2012: Towards citizen-contributed urban planning through sentiment analysis of twitter data

The dynamic nature of cities, understood as complex systems with a variety of concurring factors, poses significant challenges to urban analysis for supporting planning processes. This particularly applies to large urban events because their characteristics often contradict daily planning routines. Due to the availability of large amounts of data, social media offer the possibility for fine-scale spatial and temporal analysis in this context, especially regarding public emotions related to varied topics. Thus, this article proposes a combined approach for analyzing large sports events considering event days vs comparison days (before or after the event) and different user groups (residents vs visitors), as well as integrating sentiment analysis and topic extraction. Our results based on various analyses of tweets demonstrate that different spatial and temporal patterns can be identified, clearly distinguishing both residents and visitors, along with positive or negative sentiment. Furthermore, we could assign tweets to specific urban events or extract topics related to the transportation infrastructure. Although the results are potentially able to support urban planning processes of large events, the approach still shows some limitations including well-known biases in social media or shortcomings in identifying the user groups and in the topic modeling approach.

SCOPUS_ID:85136579190

#Lorrydeaths: Structural Topic Modeling of Twitter Users' Attitudes About the Deaths of 39 Vietnamese Migrants to the United Kingdom

In this article, we analyze anti- and pro-immigrant attitudes expressed following the Essex Lorry Deaths tragedy in October 2019 in Britain, in which 39 Vietnamese immigrants died in a sealed lorry truck on their way to their destination. We apply Structural Topic Modeling, an automated text analysis method, to a Twitter dataset (N = 4,376), to understand public responses to the Lorry Deaths incident. We find that Twitter users' posts were organized into two themes regarding attitudes toward immigrants: (1) migration narratives, stereotypes, and victim identities, and (2) border control. Within each theme, both pro- and anti-immigration attitudes were expressed. Pro-immigration posts reflected counter-narratives that challenged the mainstream media's coverage of the incident and critiqued the militarization of borders and the criminalization of immigration. Anti-immigration posts ranged from reproducing stereotypes about Vietnamese immigrants to explicitly blaming the victims themselves or their families for the deaths. This study demonstrates the uses and limitations of using Twitter for public opinion research by offering a nuanced analysis of how pro-and anti-immigration attitudes are discussed in response to a tragic event. Our research also contributes to a growing literature on public opinion about an often-forgotten immigrant group in the UK, the Vietnamese.

https://aclanthology.org//W18-0802/

#MeToo Alexa: How Conversational Systems Respond to Sexual Harassment

Conversational AI systems, such as Amazon’s Alexa, are rapidly developing from purely transactional systems to social chatbots, which can respond to a wide variety of user requests. In this article, we establish how current state-of-the-art conversational systems react to inappropriate requests, such as bullying and sexual harassment on the part of the user, by collecting and analysing the novel #MeTooAlexa corpus. Our results show that commercial systems mainly avoid answering, while rule-based chatbots show a variety of behaviours and often deflect. Data-driven systems, on the other hand, are often non-coherent, but also run the risk of being interpreted as flirtatious and sometimes react with counter-aggression. This includes our own system, trained on “clean” data, which suggests that inappropriate system behaviour is not caused by data bias.

SCOPUS_ID:85098444596

#MeToo, #MeThree, #MeFour: Twitter as community building across academic and corporate institutions

This netnography-based inquiry of the #MeToo movement on Twitter recognizes the relationships between organizations and individuals in varying levels of power and how social media can mitigate power dynamics in reporting systems for sexual harassment and assault. We study both academia and corporations as cultural institutions within which sexual harassment survivors are historically disenfranchised and convinced not to voice their stories. Distinct elements of these structures create differences in how vulnerable individuals face persecution. This paper explores the relationships between university and corporate contexts within the ecological framework of harassment. Our multidisciplinary study contributes to existing literature by extracting two samples of tweets (n = 1248 university and n = 1290 corporate) with three different unguided analytic tools to explore their semantic meaning, valence and emotionality, and overall sentiment. Drawing from our findings and literature review, we discuss the history of #MeToo as an amplification tool; Twitter as a social movement mechanism; and the roles of power, retaliation, and risk across institutions in relation to survivors of sexual violence and harassment.

SCOPUS_ID:85083712672

#MeTooMaastricht: Building a chatbot to assist survivors of sexual harassment

Inspired by the recent social movement of #MeToo, we are building a chatbot to assist survivors of sexual harassment cases (designed for the city of Maastricht but can easily be extended). The motivation behind this work is twofold: properly assist survivors of such events by directing them to appropriate institutions that can offer them help and increase the incident documentation so as to gather more data about harassment cases which are currently under reported. We break down the problem into three data science/machine learning components: harassment type identification (treated as a classification problem), spatio-temporal information extraction (treated as Named Entity Recognition problem) and dialogue with the users (treated as a slot-filling based chatbot). We are able to achieve a success rate of more than 98% for the identification of a harassment-or-not case and around 80% for the specific type harassment identification. Locations and dates are identified with more than 90% accuracy and time occurrences prove more challenging with almost 80%. Finally, initial validation of the chatbot shows great potential for the further development and deployment of such a beneficial for the whole society tool.

SCOPUS_ID:85101263699

#Metoovertising: the institutional work of creative women who are looking to change the rules of the advertising game

In the wake of #MeToo, the ad industry is coming to terms with its own issues of sexual harassment. We therefore explore the institutional work of actors who are constrained by these gendered institutional arrangements and consider how they might be involved in changing the sexist attitudes and behaviours prevalent in ad agencies. We consider the work of Les Lionnes, a collective of women working in French advertising agencies, who form a boundary organisation to address sexual harassment in the French advertising industry. By conducting critical discourse of their 2019 poster campaign, together with a netnographic study of their social media sites and an interview with its founder, we identify how advertising is used to expose the sexist attitudes and behaviours embedded in discourse and challenge the continued legitimacy of institutional logics. The success of this work may be further enhanced when it is aligned with a wider social discourse, such as #MeToo. We therefore conceptualise the advertising undertaken by Les Lionnes as institutional work which seeks to expose sexual harassment and abuse within the ad industry. We call this novel form of advertising #Metoovertising.

SCOPUS_ID:85148719377

#NotDying4Wallstreet: A Discourse Analysis on Health vs. Economy during COVID-19

This paper combines political/poststructuralist discourse theory with actor–network theory to explore dystopian visions in the context of a discourse around the hashtag #NotDying4Wallstreet. The call for protest against former US president Donald Trump’s demand to reopen the economy during lockdown dominates the discourse. The tweets were analyzed with quantitative discourse analysis and network analysis to identify key terms and meaning clusters leading to two main conclusions. The first (A) is an imaginary dystopic future with an accelerated neoliberal order. Human lives, especially elderly people, are sacrificed for a well-functioning economy in this threat scenario. The second (B) includes the motive of protest and the potential of the people’s demands to unite and rally against this threat. Due to the revelation of populist features, this (online) social movement seems to be populist without a leader figure. The empirical study is used to propose a research approach toward a mixed-methods design based on a methodological discussion and the enhancement of PDT with ANT. Thus, the article has a double aim: an update of contemporary approaches to social media analysis in discourse studies and its empirical demonstration with a study.

SCOPUS_ID:85098796046

#Populism on twitter: Statistical analysis of the correlation between tweet popularity and “populist” discursive features

Recent political events, such as the Brexit or Donald Trump's electoral success, have led to a proliferation of studies focusing on populism nature (Müller 2017; Mudde and Kaltwasser 2017). Part of the literature has also investigated communicative aspects of populism, highlighting how populists are benefitting from the use of social media (Bartlett 2014; Gerbaudo 2018). This research offers further insights on the subject by analyzing populist discourse on Twitter and exploring the correlation between the presence of linguistic features linked to populism, such as emotionalization, simplified rhetoric and intensified claims (Canovan 1999; Heinisch 2008), and tweet popularity. The use of linear mixed effects models revealed a positive correlation between the linguistic elements of interest and tweet popularity, not only in the populist sample, but also in the control group composed by establishment politicians. Surprisingly, reference tweets received more popularity than populist messages when the discursive features analyzed were present.

SCOPUS_ID:85123930393

#PraCegoVer: A Large Dataset for Image Captioning in Portuguese

Automatically describing images using natural sentences is essential to visually impaired people’s inclusion on the Internet. This problem is known as Image Captioning. There are many datasets in the literature, but most contain only English captions, whereas datasets with captions described in other languages are scarce. We introduce the #PraCegoVer, a multi-modal dataset with Portuguese captions based on posts from Instagram. It is the first large dataset for image captioning in Portuguese. In contrast to popular datasets, #PraCegoVer has only one reference per image, and both mean and variance of reference sentence length are significantly high, which makes our dataset challenging due to its linguistic aspect. We carry a detailed analysis to find the main classes and topics in our data. We compare #PraCegoVer to MS COCO dataset in terms of sentence length and word frequency. We hope that #PraCegoVer dataset encourages more works addressing the automatic generation of descriptions in Portuguese.

SCOPUS_ID:85143175521

#ROSJATOSTANUMYSLU #60KOPIEJEKZAWPIS: STEREOTYPY ETNICZNE ZAKLĘTE W HASZTAGACH

The text presents the issue of evaluating and updating ethnic stereotypes about Russia and Russians (ruski, ruscy, ruskie) implemented in internet communication using folksonomy (social categorization of content). The research area is the Polish website wykop.pl, the materials of which have been used for the contextual analysis of contemporary implementations of linguistic ethnic stereotypes manifested by objects such as tags. The operator #rosja (#russia), together with the excerpted tags, creates collocations, opening the way to renegotiating the image of stereotypes Russia and the Russian. The last part of the article presents linguistic implementations of ethnic stereotypes Russia and the Russian in tags used for discrediting and polarization in public unofficial social discourse.

http://arxiv.org/abs/1806.03369v1

#SarcasmDetection is soooo general! Towards a Domain-Independent Approach for Detecting Sarcasm

Automatic sarcasm detection methods have traditionally been designed for maximum performance on a specific domain. This poses challenges for those wishing to transfer those approaches to other existing or novel domains, which may be typified by very different language characteristics. We develop a general set of features and evaluate it under different training scenarios utilizing in-domain and/or out-of-domain training data. The best-performing scenario, training on both while employing a domain adaptation step, achieves an F1 of 0.780, which is well above baseline F1-measures of 0.515 and 0.345. We also show that the approach outperforms the best results from prior work on the same target domain.

SCOPUS_ID:85123102514

#She'sOnly16: Critical discourse analysis of posts in Twitter on a case of collective rape in Rio de Janeiro

This article investigates discourses on violence against women published on Twitter, and its interface with public policies, based on the theoretical-methodological input of Critical Discourse Analysis (CDA). The text analysis focuses on a specific case: the collective rape of a 16-year-old girl in Rio de Janeiro, Brazil, on May 21st, 2016 and the disclosure of images of this crime. The case provoked a huge uproar in Brazil. The crime became known and was reported due to the release on Twitter of a video of the rape, which was recorded by one of the attackers. This article aims at assessing the potential impact of social media on the debate of violence against women and on the responsive action of the Brazilian public authorities.

SCOPUS_ID:85127854733

#StrongTogether? Qualitative Sentiment Analysis of Social Media Reactions to Disaster Volunteering during a Forest Fire in Finland

The transformation of disaster volunteering has been highlighted in academic literature. This study examined that transformation via a big data approach. The context for the study was provided by a forest fire in Finland, which sparked a debate on volunteering. The data (806 social media messages) were analyzed using qualitative sentiment analysis to (1) identify the sentiments relating to a variety of volunteers and (2) understand the context of and tensions behind those sentiments. The data suggested that the prevailing view of disaster volunteering is a rather traditional one, while the observations on the transformation remain largely latent. The positive sentiments reflected a view of the co-production of extinguishing forest fires as an activity of formal governmental and nonprofit emergency management organizations and volunteers from expanding and extending organizations. Unaffiliated volunteers were seen as extra pairs of hands that could be invited to help in an organized way and with limited tasks, only if required. Sentiments with a more negative tone raised concerns about having sufficient numbers of affiliated volunteers in the future and the rhetorical level of appreciation of them. The data revealed a dichotomous relationship between “professionals” and “amateurs” and the politicization of the debate between different actor groups.

SCOPUS_ID:85035222514

#VisualHashtags: Visual summarization of social media events using mid-level visual elements

The data generated on social media sites continues to grow at an increasing rate with more than 36% of tweets containing images making the dominance of multimedia content evidently visible. This massive user generated content has become a reflection of world events. In order to enhance the ability and effectiveness to consume this plethora of data, summarization of these events is needed. However, very few studies have exploited the images attached with social media events to summarize them using "mid-level visual elements". These are the entities which are both representative and discriminative to the target dataset besides being human-readable and hence more informative. In this paper we propose a methodology for visual event summarization by extracting mid-level visual elements from images associated with social media events on Twitter (#VisualHashtags). The key research question is Which elements can visually capture the essence of a viral event?, hence explain its virality, and summarize it. Compared to the existing approaches of visual event summarization on social media data, we aim to discover #VisualHashtags, i.e., meaningful patches that can become the visual analog of a regular text hashtag that Twitter generates. Our algorithm incorporates a multi-stage filtering process and social popularity based ranking to discover mid-level visual elements, which overcomes the challenges faced by direct application of the existing methods. We evaluate our approach on a recently collected social media event dataset, comprising of 20,084 images. We evaluate the quality of #VisualHashtags extracted by conducting a user-centered evaluation where users are asked to rate the relevance of the resultant patches w.r.t. the event and the quality of the patch in terms of how meaningful it is. We also do a quantitative evaluation on the results. We show a high search space reduction of 93% in images and 99% in patches after summarization. Further, we get a 83% of purity in the resultant patches with a data coverage of 18%.

SCOPUS_ID:85114645945

#allforjan: How twitter users in europe reacted to the murder of ján kuciak—revealing spatiotemporal patterns through sentiment analysis and topic modeling

Social media platforms such as Twitter are considered a new mediator of collective action, in which various forms of civil movements unite around public posts, often using a common hashtag, thereby strengthening the movements. After 26 February 2018, the #AllforJan hashtag spread across the web when Ján Kuciak, a young journalist investigating corruption in Slovakia, and his fiancée were killed. The murder caused moral shock and mass protests in Slovakia and in several other European countries, as well. This paper investigates how this murder, and its follow-up events, were discussed on Twitter, in Europe, from 26 February to 15 March 2018. Our investigations, including spatiotemporal and sentiment analyses, combined with topic modeling, were conducted to comprehensively understand the trends and identify potential underlying factors in the escalation of the events. After a thorough data pre-processing including the extraction of spatial information from the users’ profile and the translation of non-English tweets, we clustered European countries based on the temporal patterns of tweeting activity in the analysis period and investigated how the sentiments of the tweets and the discussed topics varied over time in these clusters. Using this approach, we found that tweeting activity resonates not only with specific follow-up events, such as the funeral or the resignation of the Prime Minister, but in some cases, also with the political narrative of a given country affecting the course of discussions. Therefore, we argue that Twitter data serves as a unique and useful source of information for the analysis of such civil movements, as the analysis can reveal important patterns in terms of spatiotemporal and sentimental aspects, which may also help to understand protest escalation over space and time.

SCOPUS_ID:85146832725

#anxiety: A multimodal discourse analysis of narrations of anxiety on TikTok

The video-centered platform, TikTok, has gained popularity due to its position as an entertainment app, but it is still underexplored as a tool that generates awareness and discussions about mental health. This article explores TikTok's data-point ranking system to analyze how mental health rhetoric is shaped and how public health communities are formed around the term anxiety. Through a multimodal discourse analysis of the top 10 TikTok videos using the hashtag, #anxiety, this article seeks to establish how discussions of anxiety disorders are facilitated through the use of TikTok's socio-technical features and affordances of visibility, editability, persistence, and association in order to build digital communities of support. I identify recurring themes in users’ narrations of anxiety by studying in-frame content that creates meaning and contextual messages about mental health. Ultimately, these multimodal expressions of anxiety allow users to intervene and discuss often serious topics related to mental health through video, text, images, and sounds that other users can relate to and recognize. These features and affordances create networks of community and attract conversation where others can share their experiences and practices.

SCOPUS_ID:85147261119

#conspiracymemes: A framework-based analysis of conspiracy memes as digital multimodal units and ensuing user reactions on Instagram

Drawing on a multimodal corpus of Instagram memes (see Dancygier and Vandelanotte 2017; Yus 2019), this study explores the forms and functions of online conspiracy memes which are rooted in conspiracy theories (see Byford 2011; Butter and Knight 2020; Uscinski 2020). Focusing on posts, so-called digital multimodal units (DMUs), with regard to platform functions and the ensuing user reactions, this chapter qualitatively investigates DMUs with a discourse-analytic focus from two complementary angles: the posting of internet memes by the account holder as well as the ensuing meta-reflexive discussions (see Bublitz and Hübler 2007) among the audience, in which potential supporters and opponents of conspiracy theories engage to negotiate the semiotics and functions of DMUs.

SCOPUS_ID:85129744602

#facebookdown: Time to panic or detox? Understanding users' reactions to social media outage

Non-use, particularly involuntary non-use, is an under-researched topic in HCI research, even though it has become quite common nowadays due to frequent digital outages. How do users react to social media outage? Do they become anxious? Or, do they enjoy these brief episodes of social media detox? To answer these questions, we conducted a topic modeling analysis of 223,815 tweets that used the hashtag #facebookdown during the major Facebook outage on 10/4/2021. We uncovered 10 major themes of users' reactions towards social media outage. Results showed that most users complained, mocked and showed desperation about the outage situation, and during the outage period, increased their quest for other social-media alternatives. Also, surprisingly, many users celebrated the detox from Facebook rather than wishing it to come back as soon as possible. Results offer design implications for practitioners who would like to better respond to future outages.

SCOPUS_ID:85074058833

#fukushima Five Years On: A Multimethod Analysis of Twitter on the Anniversary of the Nuclear Disaster

This article examines how the fifth anniversary of the Fukushima Daiichi nuclear disaster was commemorated on English-speaking Twitter in March 2016. By combining social network analysis and critical discourse analysis, a research design is developed that can be applied to study the structure of actors and interpretative resources invoked in the crafting of communal remembrance of a disruptive, global media event. In the study, we explore the most visible actors and the most dominant meanings in the #fukushima stream. According to our analysis, the most significant players were the mainstream media and other established organizations. While most of the retweeted messages contained a ritual element of collective memory work, grief, and observance, another prominent feature was the strongly politicized discourse surrounding the aftermath of the disaster.

SCOPUS_ID:84936880891

#iamhappybecause: Gross National Happiness through Twitter analysis and big data

The prominence of social media has contributed to open information access for the researchers. With voluntary information sharing structure of Twitter, user disposition and sentiment analyses can be performed for determining the emotional well-being of the citizens. In this respect, we adopted a sentiment analysis model to calculate the Gross National Happiness (GNH) of a Middle East country, Turkey. For this purpose, over 35 million tweets, published in 2013 and in the first quarter of 2014, of over 20 thousand users were collected and analyzed. In the proposed model, prior to calculating the GNH by considering the polarities of tweets, first convergent and face validities of sentiment analysis and reliability of dataset were tested. After obtaining satisfactory results, the GNH by province survey results of Turkish Statistical Institute was compared to results of sentiment analysis for 2013 in order to state the difference between the surveying method and the proposed social media analysis method. Also, GNH by province in the first quarter of 2014 was analyzed. Additionally, relationships between users' account properties and happiness levels were investigated. Results showed that two GNH calculation approaches give similar results for the country-wide GNH levels. As a conclusion, GNH levels in the first quarter of 2014 were calculated as 47.4% happy, 28.4% neutral and 24.2% unhappy. Besides, strong correlations were found between users' happiness levels and Twitter characteristics.

SCOPUS_ID:85011945219

#naorobot: Exploring nao discourse on twitter

The Nao robot from Aldebaran Robotics is a fairly popular humanoid robot. In this study, we aimed to conduct a discourse analysis around the Nao through content analysis of posts on Twitter (N=235 tweets in English). The analysis aimed to understand discourse around the robot, the usage trends of the robot, the existence of social relationships between user and robot and if there were any patterns in tweeting. Our main results show that the Nao is attributed with high anthropomorphism and social rapport and the popular usages of the robot extend to research and education but not to health and domestic applications. In conclusion, we speculate on our results obtained and present a direction for future research.

SCOPUS_ID:85064704542

#secondcivilwarletters from the front: Discursive illusions in a trending Twitter hashtag

When conservative media personality and conspiracy theorist Alex Jones warned of an impending Second Civil War to be initiated by Democrats, he instigated a viral hashtag on Twitter – #secondcivilwarletters – which drew tweets of political commentary and critique in a style mimicking war letters from the American Civil War. Using a sample of these tweets, this article explores the evocation of discursive illusions already established within mainstream and alternative media discourse about contemporary partisan politics in America – that is, the divide between the Republicans and the Democrats and how they categorise each other. To do this we adopt Bhatia’s framework of the discourse of illusion and its three main components of linguistic and semiotic action, historicity and social impact. The analysis reveals the extent to which this illusion has permeated the consciousness of the users as they present their ideological beliefs and positions in this new media context.

SCOPUS_ID:85075460752

#suicidal - A multipronged approach to identify and explore suicidal ideation in twitter

Technological advancements have led to the creation of social media platforms like Twitter, where people have started voicing their views over rarely discussed and socially stigmatizing issues. Twitter, is increasingly being used for studying psycho-linguistic phenomenon spanning from expressions of adverse drug reactions, depressions, to suicidality. In this work we focus on identifying suicidal posts from Twitter. Towards this objective we take a multipronged approach and implement different neural network models such as sequential models and graph convolutional networks, that are trained on textual content shared in Twitter, the historical tweeting activity of the users and social network formed between different users posting about suicidality. We train a stacked ensemble of classifiers representing different aspects of suicidal tweeting activity, and achieve state-of-the-art results on a new manually annotated dataset developed by us, that contains textual as well as network information of suicidal tweets. We further investigate into the trained models and perform qualitative analysis showing how historical tweeting activity and rich information embedded in the homophily networks amongst users in Twitter, aids in accurately identifying tweets expressing suicidal intent.

SCOPUS_ID:85090246152

#sustainablefashion–A Conceptual Framework for Sustainable Fashion Discourse on Twitter

The concept of sustainability has gained in importance since the United Nations made it the main pillar for development in 1992. But while areas like climate change and energy consumption are well known, the concept of sustainable fashion is neither understood nor conceptualized very well. Most research in the field focuses on fashion consumption but has various understandings of sustainability. Through a text-based analysis of the Twitter discourse on sustainable fashion, this study developed a conceptual framework to understand perceptions of sustainable fashion that can help guide theory and practice. The findings contribute to the development of a circular model of sustainable fashion that combines production, diffusion, and consumption processes. The utility and limitations of the conceptual model are discussed.

SCOPUS_ID:85096919583

#welcomerefugees: A critical discourse analysis of the refugee resettlement initiative in Canadian news

This study focuses on the frames utilized in the depiction of Syrian refugees and social and political actors involved in the Syrian resettlement in Canadian online news media. The role of the media is vital in portraying Syrian refugees' image and affects how the Canadian public perceives them. This paper focuses on utilizing the referential and predicational strategies introduced by the Discourse-Historical Approach (DHA) in framing the Syrian refugees, Liberal government, Conservative party, Canadians, and Canada (henceforth social and political actors). This study examines a total of 31 articles selected from three of the most visited Canadian news sites, namely, the Toronto Star, the Toronto Sun, and the National Post. News articles were collected beginning from the arrival of the first group of refugees in December 2015 and ending in March 2017, which marked the first anniversary of the refugees’ arrival. The results obtained show that both liberal and conservative-leaning media utilized frames in ways that correspond with their ideological stance. In most cases, the limelight rarely focused on Syrian refugees. Instead, they were used as props to push the news source's ideological convictions and to condemn and shame the opposition. Therefore, it is understood, that the framing and portrayal of refugees in this narrow manner through discursive strategies obscures the complexity of the plight of Syrian refugees and depicts them as one-dimensional characters that audiences would either fear or pity.

http://arxiv.org/abs/1711.01921v3

$A^{4}NT$: Author Attribute Anonymity by Adversarial Training of Neural Machine Translation

Text-based analysis methods allow to reveal privacy relevant author attributes such as gender, age and identify of the text's author. Such methods can compromise the privacy of an anonymous author even when the author tries to remove privacy sensitive content. In this paper, we propose an automatic method, called Adversarial Author Attribute Anonymity Neural Translation ($A^4NT$), to combat such text-based adversaries. We combine sequence-to-sequence language models used in machine translation and generative adversarial networks to obfuscate author attributes. Unlike machine translation techniques which need paired data, our method can be trained on unpaired corpora of text containing different authors. Importantly, we propose and evaluate techniques to impose constraints on our $A^4NT$ to preserve the semantics of the input text. $A^4NT$ learns to make minimal changes to the input text to successfully fool author attribute classifiers, while aiming to maintain the meaning of the input. We show through experiments on two different datasets and three settings that our proposed method is effective in fooling the author attribute classifiers and thereby improving the anonymity of authors.

http://arxiv.org/abs/2106.08914v1

$C^3$: Compositional Counterfactual Constrastive Learning for Video-grounded Dialogues

Video-grounded dialogue systems aim to integrate video understanding and dialogue understanding to generate responses that are relevant to both the dialogue and video context. Most existing approaches employ deep learning models and have achieved remarkable performance, given the relatively small datasets available. However, the results are partly accomplished by exploiting biases in the datasets rather than developing multimodal reasoning, resulting in limited generalization. In this paper, we propose a novel approach of Compositional Counterfactual Contrastive Learning ($C^3$) to develop contrastive training between factual and counterfactual samples in video-grounded dialogues. Specifically, we design factual/counterfactual sampling based on the temporal steps in videos and tokens in dialogues and propose contrastive loss functions that exploit object-level or action-level variance. Different from prior approaches, we focus on contrastive hidden state representations among compositional output tokens to optimize the representation space in a generation setting. We achieved promising performance gains on the Audio-Visual Scene-Aware Dialogues (AVSD) benchmark and showed the benefits of our approach in grounding video and dialogue context.

http://arxiv.org/abs/2302.01328v2

$IC^3$: Image Captioning by Committee Consensus

If you ask a human to describe an image, they might do so in a thousand different ways. Traditionally, image captioning models are trained to approximate the reference distribution of image captions, however, doing so encourages captions that are viewpoint-impoverished. Such captions often focus on only a subset of the possible details, while ignoring potentially useful information in the scene. In this work, we introduce a simple, yet novel, method: "Image Captioning by Committee Consensus" ($IC^3$), designed to generate a single caption that captures high-level details from several viewpoints. Notably, humans rate captions produced by $IC^3$ at least as helpful as baseline SOTA models more than two thirds of the time, and $IC^3$ captions can improve the performance of SOTA automated recall systems by up to 84%, indicating significant material improvements over existing SOTA approaches for visual description. Our code is publicly available at https://github.com/DavidMChan/caption-by-committee

http://arxiv.org/abs/2210.14431v3

$N$-gram Is Back: Residual Learning of Neural Text Generation with $n$-gram Language Model

$N$-gram language models (LM) have been largely superseded by neural LMs as the latter exhibits better performance. However, we find that $n$-gram models can achieve satisfactory performance on a large proportion of testing cases, indicating they have already captured abundant knowledge of the language with relatively low computational cost. With this observation, we propose to learn a neural LM that fits the residual between an $n$-gram LM and the real-data distribution. The combination of $n$-gram and neural LMs not only allows the neural part to focus on the deeper understanding of language but also provides a flexible way to customize an LM by switching the underlying $n$-gram model without changing the neural model. Experimental results on three typical language tasks (i.e., language modeling, machine translation, and summarization) demonstrate that our approach attains additional performance gains over popular standalone neural models consistently. We also show that our approach allows for effective domain adaptation by simply switching to a domain-specific $n$-gram model, without any extra training. Our code is released at https://github.com/ghrua/NgramRes.

http://arxiv.org/abs/2303.09522v1

$P+$: Extended Textual Conditioning in Text-to-Image Generation

We introduce an Extended Textual Conditioning space in text-to-image models, referred to as $P+$. This space consists of multiple textual conditions, derived from per-layer prompts, each corresponding to a layer of the denoising U-net of the diffusion model. We show that the extended space provides greater disentangling and control over image synthesis. We further introduce Extended Textual Inversion (XTI), where the images are inverted into $P+$, and represented by per-layer tokens. We show that XTI is more expressive and precise, and converges faster than the original Textual Inversion (TI) space. The extended inversion method does not involve any noticeable trade-off between reconstruction and editability and induces more regular inversions. We conduct a series of extensive experiments to analyze and understand the properties of the new space, and to showcase the effectiveness of our method for personalizing text-to-image models. Furthermore, we utilize the unique properties of this space to achieve previously unattainable results in object-style mixing using text-to-image models. Project page: https://prompt-plus.github.io

http://arxiv.org/abs/2104.08202v2

$Q^{2}$: Evaluating Factual Consistency in Knowledge-Grounded Dialogues via Question Generation and Question Answering

Neural knowledge-grounded generative models for dialogue often produce content that is factually inconsistent with the knowledge they rely on, making them unreliable and limiting their applicability. Inspired by recent work on evaluating factual consistency in abstractive summarization, we propose an automatic evaluation metric for factual consistency in knowledge-grounded dialogue using automatic question generation and question answering. Our metric, denoted $Q^2$, compares answer spans using natural language inference (NLI), instead of token-based matching as done in previous work. To foster proper evaluation, we curate a novel dataset of dialogue system outputs for the Wizard-of-Wikipedia dataset, manually annotated for factual consistency. We perform a thorough meta-evaluation of $Q^2$ against other metrics using this dataset and two others, where it consistently shows higher correlation with human judgements.

http://arxiv.org/abs/2004.13248v4

$R^3$: Reverse, Retrieve, and Rank for Sarcasm Generation with Commonsense Knowledge

We propose an unsupervised approach for sarcasm generation based on a non-sarcastic input sentence. Our method employs a retrieve-and-edit framework to instantiate two major characteristics of sarcasm: reversal of valence and semantic incongruity with the context which could include shared commonsense or world knowledge between the speaker and the listener. While prior works on sarcasm generation predominantly focus on context incongruity, we show that combining valence reversal and semantic incongruity based on the commonsense knowledge generates sarcasm of higher quality. Human evaluation shows that our system generates sarcasm better than human annotators 34% of the time, and better than a reinforced hybrid baseline 90% of the time.

http://arxiv.org/abs/2109.00301v3

$\infty$-former: Infinite Memory Transformer

Transformers are unable to model long-term memories effectively, since the amount of computation they need to perform grows with the context length. While variations of efficient transformers have been proposed, they all have a finite memory capacity and are forced to drop old information. In this paper, we propose the $\infty$-former, which extends the vanilla transformer with an unbounded long-term memory. By making use of a continuous-space attention mechanism to attend over the long-term memory, the $\infty$-former's attention complexity becomes independent of the context length, trading off memory length with precision. In order to control where precision is more important, $\infty$-former maintains "sticky memories" being able to model arbitrarily long contexts while keeping the computation budget fixed. Experiments on a synthetic sorting task, language modeling, and document grounded dialogue generation demonstrate the $\infty$-former's ability to retain information from long sequences.

http://arxiv.org/abs/2202.09817v2

$\mathcal{Y}$-Tuning: An Efficient Tuning Paradigm for Large-Scale Pre-Trained Models via Label Representation Learning

With the success of large-scale pre-trained models (PTMs), how efficiently adapting PTMs to downstream tasks has attracted tremendous attention, especially for PTMs with billions of parameters. Although some parameter-efficient tuning paradigms have been proposed to address this problem, they still require large resources to compute the gradients in the training phase. In this paper, we propose $\mathcal{Y}$-Tuning, an efficient yet effective paradigm to adapt frozen large-scale PTMs to specific downstream tasks. $\mathcal{Y}$-tuning learns dense representations for labels $\mathcal{Y}$ defined in a given task and aligns them to fixed feature representation. Without tuning the features of input text and model parameters, $\mathcal{Y}$-tuning is both parameter-efficient and training-efficient. For $\text{DeBERTa}_\text{XXL}$ with 1.6 billion parameters, $\mathcal{Y}$-tuning achieves performance more than $96\%$ of full fine-tuning on GLUE Benchmark with only $2\%$ tunable parameters and much fewer training costs.

http://arxiv.org/abs/2202.07880v4

$\rm{C {\small IS}}^2$: A Simplified Commonsense Inference Evaluation for Story Prose

Transformers have been showing near-human performance on a variety of tasks, but they are not without their limitations. We discuss the issue of conflating results of transformers that are instructed to do multiple tasks simultaneously. In particular, we focus on the domain of commonsense reasoning within story prose, which we call contextual commonsense inference (CCI). We look at the GLUCOSE (Mostafazadeh et al. 2020) dataset and task for predicting implicit commonsense inferences between story sentences. Since the GLUCOSE task simultaneously generates sentences and predicts the CCI relation, there is a conflation in the results. Is the model really measuring CCI or is its ability to generate grammatical text carrying the results? In this paper, we introduce the task contextual commonsense inference in sentence selection ($\rm{C {\small IS}}^2$), a simplified task that avoids conflation by eliminating language generation altogether. Our findings emphasize the necessity of future work to disentangle language generation from the desired NLP tasks at hand.

http://arxiv.org/abs/2204.02030v1

$\textit{latent}$-GLAT: Glancing at Latent Variables for Parallel Text Generation

Recently, parallel text generation has received widespread attention due to its success in generation efficiency. Although many advanced techniques are proposed to improve its generation quality, they still need the help of an autoregressive model for training to overcome the one-to-many multi-modal phenomenon in the dataset, limiting their applications. In this paper, we propose $\textit{latent}$-GLAT, which employs the discrete latent variables to capture word categorical information and invoke an advanced curriculum learning technique, alleviating the multi-modality problem. Experiment results show that our method outperforms strong baselines without the help of an autoregressive model, which further broadens the application scenarios of the parallel decoding paradigm.

http://arxiv.org/abs/1708.07863v1

$k$-Nearest Neighbor Augmented Neural Networks for Text Classification

In recent years, many deep-learning based models are proposed for text classification. This kind of models well fits the training set from the statistical point of view. However, it lacks the capacity of utilizing instance-level information from individual instances in the training set. In this work, we propose to enhance neural network models by allowing them to leverage information from $k$-nearest neighbor (kNN) of the input text. Our model employs a neural network that encodes texts into text embeddings. Moreover, we also utilize $k$-nearest neighbor of the input text as an external memory, and utilize it to capture instance-level information from the training set. The final prediction is made based on features from both the neural network encoder and the kNN memory. Experimental results on several standard benchmark datasets show that our model outperforms the baseline model on all the datasets, and it even beats a very deep neural network model (with 29 layers) in several datasets. Our model also shows superior performance when training instances are scarce, and when the training set is severely unbalanced. Our model also leverages techniques such as semi-supervised training and transfer learning quite well.

http://arxiv.org/abs/2302.10879v1

$k$NN-Adapter: Efficient Domain Adaptation for Black-Box Language Models

Fine-tuning a language model on a new domain is standard practice for domain adaptation. However, it can be infeasible when it comes to modern large-scale language models such as GPT-3, which can only be accessed through APIs, making it difficult to access the internal parameters of the model. In this paper, we propose $k$NN-Adapter, a method to effectively adapt these black-box large language models (LLMs) to a new domain. The $k$NN-Adapter builds on top of the retrieval-augmented language model, and adaptively learns to interpolate the output of the language model with retrieval results from a datastore consisting of the target domain data. Our experiments on four different domains demonstrate that $k$NN-Adapter significantly improves perplexity, and works particularly well in settings with limited access to LLMs. Additionally, we show that $k$NN-Adapter is more effective than fine-tuning when the amount of training data is limited. We also release a dataset to encourage further study.

http://arxiv.org/abs/2203.17103v1

$k$NN-NER: Named Entity Recognition with Nearest Neighbor Search

Inspired by recent advances in retrieval augmented methods in NLP~\citep{khandelwal2019generalization,khandelwal2020nearest,meng2021gnn}, in this paper, we introduce a $k$ nearest neighbor NER ($k$NN-NER) framework, which augments the distribution of entity labels by assigning $k$ nearest neighbors retrieved from the training set. This strategy makes the model more capable of handling long-tail cases, along with better few-shot learning abilities. $k$NN-NER requires no additional operation during the training phase, and by interpolating $k$ nearest neighbors search into the vanilla NER model, $k$NN-NER consistently outperforms its vanilla counterparts: we achieve a new state-of-the-art F1-score of 72.03 (+1.25) on the Chinese Weibo dataset and improved results on a variety of widely used NER benchmarks. Additionally, we show that $k$NN-NER can achieve comparable results to the vanilla NER model with 40\% less amount of training data. Code available at \url{https://github.com/ShannonAI/KNN-NER}.

http://arxiv.org/abs/2210.11912v1

$m^4Adapter$: Multilingual Multi-Domain Adaptation for Machine Translation with a Meta-Adapter

Multilingual neural machine translation models (MNMT) yield state-of-the-art performance when evaluated on data from a domain and language pair seen at training time. However, when a MNMT model is used to translate under domain shift or to a new language pair, performance drops dramatically. We consider a very challenging scenario: adapting the MNMT model both to a new domain and to a new language pair at the same time. In this paper, we propose $m^4Adapter$ (Multilingual Multi-Domain Adaptation for Machine Translation with a Meta-Adapter), which combines domain and language knowledge using meta-learning with adapters. We present results showing that our approach is a parameter-efficient solution which effectively adapts a model to both a new language pair and a new domain, while outperforming other adapter methods. An ablation study also shows that our approach more effectively transfers domain knowledge across different languages and language information across different domains.

SCOPUS_ID:79956233129

$urplus: Spinoza, Lacan, by A. Kiarina Kordela. Albany: State university of New York press, 2007

In this review I trace Kordela's original and brilliantly argued claim that the Spinozian-Marxian line of thought finds its proper articulation in contemporary Lacanian psychoanalysis. I illuminate the linkages and logic underpinning her analysis, including her unique and complex reading of Spinoza's monism, which yields critical insights into paradoxes of belief, truth, and causality; her meticulous counterarguments to Alain Badiou, Slavoj Žižek, Gilles Deleuze, and Michael Hardt and Antonio Negri, among others; her extension of Kojin Karatani's work on set theory in Marx by way of Lacan's concepts of the gaze and feminine ethics; and her critique of Immanuel Kant's antinomies as they relate to global capitalism. Ultimately, I highlight where Kordela's claims lead, both in terms of philosophy (the obsolescence of Platonism) and in terms of a new pathway for a revisionist conception of Marxism as a theory of language, being, and ethics. © 2010 Association for Economic and Social Analysis.