Automatic labelling of topic models learned from Twitter by summarisation

Latent topics derived by topic models such as Latent Dirichlet Allocation (LDA) are the result of hidden thematic structures which provide further insights into the data. The automatic labelling of such topics derived from social media poses however new challenges since topics may characterise novel events happening in the real world. Existing automatic topic labelling approaches which depend on external knowledge sources become less applicable here since relevant articles/concepts of the extracted topics may not exist in external sources. In this paper we propose to address the problem of automatic labelling of latent topics learned from Twitter as a summarisation problem. We introduce a framework which apply summarisation algorithms to generate topic labels. These algorithms are independent of external sources and only rely on the identification of dominant terms in documents related to the latent topic. We compare the efficiency of existing state of the art summarisation algorithms. Our results suggest that summarisation algorithms generate better topic labels which capture event-related context compared to the top-n terms returned by LDA. © 2014 Association for Computational Linguistics.

'If you had only listened carefully...':the discursive construction of emerging leadership in a UK all-women management team

Increasingly, feminist linguistic research has adopted a discursive perspective to learn how women and men 'do' leadership in gendered ways. 'Women' as a social category is made relevant to this study by virtue of the lack of female senior leaders in UK businesses (Sealy and Vinnicombe, 2013). Much previous research has analysed leadership discourse in mixed gender groups, relying on theories that imply comparisons between men and women. Using an Interactional Sociolinguistic approach, this study aims to learn more about how women perform leadership in the absence of men by analysing the spoken interactions of a women-only team who were engaged in a competitive leadership task. The analysis reveals that the women accomplish leadership in multiple and complex ways that defy binary gendered classifications. Nonetheless, there is a distinctive gendered dynamic to the team's interactions which, it is argued, might be disadvantageous to women aspiring to senior positions. © The Author(s) 2013.

'Your situation is critical...':the discursive enactment of leadership by business women in Middle Eastern and Western European contexts

Women remain in a small minority as business leaders in both Middle Eastern (ME) and Western European (WE) regions, and indeed, past research indicates that ME women face even greater challenges as leaders than their Western counterparts. This article explores sample findings from two separate case studies, the first of a ME woman leader and the second of a WE woman leader, each conducting a management meeting with their teams. Using interactional sociolinguistic analysis, we examine the 'contextualisation cues' that index how each woman performs leadership in their respective meetings. We found that both women utilise relational practices in order to enact leadership with their subordinates, but with varying results. Whereas the ME leader deploys a confident and commanding interactional style with her colleagues, the WE leader's style is evasive and uncertain. On the basis of these two cases, the WE leader appears to face greater challenges in a male-dominated business world than the ME leader. Whereas the ME leader can rely on long-established ties of loyalty and organisation-as-family, the Western leader, within an apparently more open, democratic context, has to negotiate overwhelming turbulence and change within her company. © 2014, equinox publishing.

E-HR and international HRM:a critical perspective on the discursive framing of e-HR

In the current global economic climate, international HRM is facing unprecedented pressure to become more innovative, effective and efficient. New discourses are emerging around the application of information technology, with 'e-HR' (electronic-enablement of Human Resources), self-service portals and promises of improved services couched as various HR 'value propositions'. This study explores these issues through our engagement with the emergent stream of 'critical' HRM, the broader study of organizational discourse and ethical management theories. We have found that while there is growing research into the take-up of e-HR applications, there is a dearth of investigation into the impact of e-HR on the people involved; in particular, the (re)structuring of social relations between HR functions and line managers in the move away from face-to-face HR support services, to more technology-mediated 'self-service' relationships. We undertake a close reading of personal narratives from a multinational organization, deploying a critical discourse lens to examine different dimensions of e-HR and raise questions about the strong technocratic framing of the international language of people management, shaping line manager enactment of e-HR duties. We argue for a more reflexive stance in the conceptualization e-HR, and conclude with a discussion about the theoretical and practical implications of our study, limitations and suggestions for future research. © 2014 © 2014 Taylor & Francis.

Dynamic joint sentiment-topic model

Social media data are produced continuously by a large and uncontrolled number of users. The dynamic nature of such data requires the sentiment and topic analysis model to be also dynamically updated, capturing the most recent language use of sentiments and topics in text. We propose a dynamic Joint Sentiment-Topic model (dJST) which allows the detection and tracking of views of current and recurrent interests and shifts in topic and sentiment. Both topic and sentiment dynamics are captured by assuming that the current sentiment-topic-specific word distributions are generated according to the word distributions at previous epochs. We study three different ways of accounting for such dependency information: (1) Sliding window where the current sentiment-topic word distributions are dependent on the previous sentiment-topic-specific word distributions in the last S epochs; (2) skip model where history sentiment topic word distributions are considered by skipping some epochs in between; and (3) multiscale model where previous long- and shorttimescale distributions are taken into consideration. We derive efficient online inference procedures to sequentially update the model with newly arrived data and show the effectiveness of our proposed model on the Mozilla add-on reviews crawled between 2007 and 2011. © 2013 ACM 2157-6904/2013/12-ART5 $ 15.00.

Who wants to be the leader? The discursive construction of emerging leadership in differently gendered teams

Increasingly, scholars are contesting the value of grand theories of leadership in favour of a social constructionist approach that posits the centrality of language for ‘doing’ leadership. This article investigates the extent to which the linguistic enactment of leadership is often gendered, which may have consequences for the career progression of women business leaders. Drawing on a UK-based study of three teams with different gender compositions (men-only; women-only and mixed gender), I use an Interactional Sociolinguistic framework to compare what leadership ‘looks and sounds like’ during the course of a competitive, leadership task. My findings show that the linguistic construction of leadership varies considerably within each team although not always in conventionally gendered ways. The study potentially provides linguistic insights on the business issue of why so few women progress from middle management to senior leadership roles.

Incorporating social role theory into topic models for social media content analysis

In this paper, we explore the idea of social role theory (SRT) and propose a novel regularized topic model which incorporates SRT into the generative process of social media content. We assume that a user can play multiple social roles, and each social role serves to fulfil different duties and is associated with a role-driven distribution over latent topics. In particular, we focus on social roles corresponding to the most common social activities on social networks. Our model is instantiated on microblogs, i.e., Twitter and community question-answering (cQA), i.e., Yahoo! Answers, where social roles on Twitter include "originators" and "propagators", and roles on cQA are "askers" and "answerers". Both explicit and implicit interactions between users are taken into account and modeled as regularization factors. To evaluate the performance of our proposed method, we have conducted extensive experiments on two Twitter datasets and two cQA datasets. Furthermore, we also consider multi-role modeling for scientific papers where an author's research expertise area is considered as a social role. A novel application of detecting users' research interests through topical keyword labeling based on the results of our multi-role model has been presented. The evaluation results have shown the feasibility and effectiveness of our model.

From Spanish paintings to murder:topic transitions in casual conversations between native and non-native speakers of English

This article explores some of the strategies used by international students of English to manage topic shifts in casual conversations with English-speaking peers. It therefore covers aspects of discourse which have been comparatively under-researched, and where research has also tended to focus on the problems rather than the communicative achievements of non-native speakers. A detailed analysis of the conversations under discussion, which were recorded by the participants themselves, showed that they all flowed smoothly, and this was in large measure due to the ways in which topic shifts were managed. The paper will focus on a very distinct type of topic shift, namely that of topic transitions, which enable a smooth flow from one topic to another, but which do not explicitly signal that a shift is taking place. It will examine how the non-native speakers achieved coherence in the topic transitions which they initiated, which strategies or procedures they employed, and show how their initiations were effective in enabling the proposed topic to be understood, taken up and developed. It therefore adds to our understanding of the interactional achievements of international speakers in informal, social contexts. © 2013 Elsevier B.V.

From art for arts sake to art as means of knowing:a rationale for advancing arts-based methods in research, practice and pedagogy

This paper advances a philosophically informed rationale for the broader, reflexive and practical application of arts-based methods to benefit research, practice and pedagogy. It addresses the complexity and diversity of learning and knowing, foregrounding a cohabitative position and recognition of a plurality of research approaches, tailored and responsive to context. Appreciation of art and aesthetic experience is situated in the everyday, underpinned by multi-layered exemplars of pragmatic visual-arts narrative inquiry undertaken in the third, creative and communications sectors. Discussion considers semi-guided use of arts-based methods as a conduit for topic engagement, reflection and intersubjective agreement; alongside observation and interpretation of organically employed approaches used by participants within daily norms. Techniques span handcrafted (drawing), digital (photography), hybrid (cartooning), performance dimensions (improvised installations) and music (metaphor and structure). The process of creation, the artefact/outcome produced and experiences of consummation are all significant, with specific reflexivity impacts. Exploring methodology and epistemology, both the "doing" and its interpretation are explicated to inform method selection, replication, utility, evaluation and development of cross-media skills literacy. Approaches are found engaging, accessible and empowering, with nuanced capabilities to alter relationships with phenomena, experiences and people. By building a discursive space that reduces barriers; emancipation, interaction, polyphony, letting-go and the progressive unfolding of thoughts are supported, benefiting ways of knowing, narrative (re)construction, sensory perception and capacities to act. This can also present underexplored researcher risks in respect to emotion work, self-disclosure, identity and agenda. The paper therefore elucidates complex, intricate relationships between form and content, the represented and the representation or performance, researcher and participant, and the self and other. This benefits understanding of phenomena including personal experience, sensitive issues, empowerment, identity, transition and liminality. Observations are relevant to qualitative and mixed methods researchers and a multidisciplinary audience, with explicit identification of challenges, opportunities and implications.

Linked knowledge sources for topic classification of microposts:a semantic graph-based approach

Short text messages a.k.a Microposts (e.g. Tweets) have proven to be an effective channel for revealing information about trends and events, ranging from those related to Disaster (e.g. hurricane Sandy) to those related to Violence (e.g. Egyptian revolution). Being informed about such events as they occur could be extremely important to authorities and emergency professionals by allowing such parties to immediately respond. In this work we study the problem of topic classification (TC) of Microposts, which aims to automatically classify short messages based on the subject(s) discussed in them. The accurate TC of Microposts however is a challenging task since the limited number of tokens in a post often implies a lack of sufficient contextual information. In order to provide contextual information to Microposts, we present and evaluate several graph structures surrounding concepts present in linked knowledge sources (KSs). Traditional TC techniques enrich the content of Microposts with features extracted only from the Microposts content. In contrast our approach relies on the generation of different weighted semantic meta-graphs extracted from linked KSs. We introduce a new semantic graph, called category meta-graph. This novel meta-graph provides a more fine grained categorisation of concepts providing a set of novel semantic features. Our findings show that such category meta-graph features effectively improve the performance of a topic classifier of Microposts. Furthermore our goal is also to understand which semantic feature contributes to the performance of a topic classifier. For this reason we propose an approach for automatic estimation of accuracy loss of a topic classifier on new, unseen Microposts. We introduce and evaluate novel topic similarity measures, which capture the similarity between the KS documents and Microposts at a conceptual level, considering the enriched representation of these documents. Extensive evaluation in the context of Emergency Response (ER) and Violence Detection (VD) revealed that our approach outperforms previous approaches using single KS without linked data and Twitter data only up to 31.4% in terms of F1 measure. Our main findings indicate that the new category graph contains useful information for TC and achieves comparable results to previously used semantic graphs. Furthermore our results also indicate that the accuracy of a topic classifier can be accurately predicted using the enhanced text representation, outperforming previous approaches considering content-based similarity measures. © 2014 Elsevier B.V. All rights reserved.

The discursive construction of evidence in police interviews:case study of a rape suspect

This article presents an analysis of the discursive construction of evidence in an English police interview with a rape suspect. The analytic findings differ from previous research on police–suspect interview discourse, in that here the interviewers actively lead an interviewee to produce defence evidence. The article seeks to make the following contributions: (i) it demonstrates the interactional mechanisms through which the interviewers co-construct the interviewee’s own version of events, and highlights the potential legal ramifications by focusing on the construction of one key evidential aspect, namely, consent; (ii) it lends weight to the hypothesis that interviewer agendas are strongly determinative of interview outcomes in terms of the evidential account produced, while making the important new contribution of showing that this is not simply a case of police interviewers being inevitably prosecution-focused; and (iii) it aims to provoke further investigation into the significance of interviewer discursive influence in cases where consent is at issue, against a backdrop of increasing numbers of rape cases being discontinued by the police at this early stage of the criminal justice process.

Harnessing linked knowledge sources for topic classification in social media

Topic classification (TC) of short text messages offers an effective and fast way to reveal events happening around the world ranging from those related to Disaster (e.g. Sandy hurricane) to those related to Violence (e.g. Egypt revolution). Previous approaches to TC have mostly focused on exploiting individual knowledge sources (KS) (e.g. DBpedia or Freebase) without considering the graph structures that surround concepts present in KSs when detecting the topics of Tweets. In this paper we introduce a novel approach for harnessing such graph structures from multiple linked KSs, by: (i) building a conceptual representation of the KSs, (ii) leveraging contextual information about concepts by exploiting semantic concept graphs, and (iii) providing a principled way for the combination of KSs. Experiments evaluating our TC classifier in the context of Violence detection (VD) and Emergency Responses (ER) show promising results that significantly outperform various baseline models including an approach using a single KS without linked data and an approach using only Tweets. Copyright 2013 ACM.

Negotiating topic changes:native and non-native speakers of English in conversation

There is a tendency to view conversations involving non-native speakers (NNSs) as inevitably fraught with problems, including an inability to handle topic management. This article, in contrast, will focus on effective topic changes made by non-native speakers during informal conversations with native speakers of English. A micro-analysis of ten conversations revealed several ways of shifting conversational topics; however, the article concentrates on those strategies which the participants used to effect a particular type of topic move, namely 'marked topic changes', where there is no connection at all with previous talk. The findings show how these topic changes were jointly negotiated, and that the non-native speakers' contributions to initiating new topics were competently managed.

Topic extraction from microblog posts using conversation structures

Conventional topic models are ineffective for topic extraction from microblog messages since the lack of structure and context among the posts renders poor message-level word co-occurrence patterns. In this work, we organize microblog posts as conversation trees based on reposting and replying relations, which enrich context information to alleviate data sparseness. Our model generates words according to topic dependencies derived from the conversation structures. In specific, we differentiate messages as leader messages, which initiate key aspects of previously focused topics or shift the focus to different topics, and follower messages that do not introduce any new information but simply echo topics from the messages that they repost or reply. Our model captures the different extents that leader and follower messages may contain the key topical words, thus further enhances the quality of the induced topics. The results of thorough experiments demonstrate the effectiveness of our proposed model.

Discursive approaches to language policy

This book brings together the fields of language policy and discourse studies from a multidisciplinary theoretical, methodological and empirical perspective. The chapters in this volume are written by international scholars active in the field of language policy and planning and discourse studies. The diverse research contexts range from education in Paraguay and Luxembourg via businesses in Wales to regional English language policies in Tajikistan. Readers are thereby invited to think critically about the mutual relationship between language policy and discourse in a range of social, political, economic and cultural spheres. Using approaches that draw on discourse-analytic, anthropological, ethnographic and critical sociolinguistic frameworks, the contributors in this collection explore and refine the ‘discursive’ and the ‘critical’ aspects of language policy as a multilayered, fluid, ideological, discursive and social process that can operate as a tool of social change as well as reinforcing established power structures and inequalities.