Visual Text Analytics Research Articles

Non-Governmental Organizations as a Place for Cultural Rituals: Cyprus Turkish French Cultural Association

This study reveals the strength non-governmental organizations as a place for recalling cultural memory and cultural rituals. In this study, the theories about the tightness of cultural memory to its cultural environment, and the non-governmental organizations through the cross-border travel and the possibility of blending with the existing cultural symbols are described. The Turkish Cypriot French Cultural Association, which has been active in Northern Cyprus for 34 years, has been presented with a historical view, but the symbolic cultural rituals of the living legacies of French culture have been addressed through civil society organizations. While the subjects of cultural memory and collective memory are often referred to politics, non-governmental organizations with contributions to culture are often ignored. In the light of these conceptual debates a wide range of cultural memory and symbolic cultural rituals have been discussed. As a concrete assessment of all these information and discussions, the Turkish Cypriot French Cultural Association's French symbolic cultural rituals are the main objective of the study. In this context, the analysis of French cultural symbolic rituals in Northern Cyprus has been demonstrated by using visual text analysis technique and content-analysis technique. As the information and documents are obtained, the Turkish Cypriot French Cultural Association as a well-established non-governmental organization in Northern Cyprus has demonstrated its importance as a cultural memory venue.

Discursive activism in the Russian feminist hashtag campaign: the #ItIsNotAReasonToKill case

ABSTRACTThis paper describes a study of the feminist hashtag campaign against blaming victims #этонеповодубить (it is not a reason to kill), which was launched in the Russian social media in 2018. This article considers #этонеповодубить as a case of discursive activism and combines techniques of discourse analysis with network analysis to reveal how the participants in the temporary community were opposed to the victim-blaming discourse. The strategies of legitimization that the participants used to justify their opinions and the provocative style of the campaign were investigated based on Reyes’ classification. Visual network analysis and text analysis were facilitated by Netlytic.

Bridging Text Visualization and Mining: A Task-Driven Survey.

Visual text analytics has recently emerged as one of the most prominent topics in both academic research and the commercial world. To provide an overview of the relevant techniques and analysis tasks, as well as the relationships between them, we comprehensively analyzed 263 visualization papers and 4,346 mining papers published between 1992-2017 in two fields: visualization and text mining. From the analysis, we derived around 300 concepts (visualization techniques, mining techniques, and analysis tasks) and built a taxonomy for each type of concept. The co-occurrence relationships between the concepts were also extracted. Our research can be used as a stepping-stone for other researchers to 1) understand a common set of concepts used in this research topic; 2) facilitate the exploration of the relationships between visualization techniques, mining techniques, and analysis tasks; 3) understand the current practice in developing visual text analytics tools; 4) seek potential research opportunities by narrowing the gulf between visualization and mining techniques based on the analysis tasks; and 5) analyze other interdisciplinary research areas in a similar way. We have also contributed a web-based visualization tool for analyzing and understanding research trends and opportunities in visual text analytics.

Interactive topic hierarchy revision for exploring a collection of online conversations

In the last decade, there has been an exponential growth of asynchronous online conversations (e.g. blogs), thanks to the rise of social media. Analyzing and gaining insights from such discussions can be quite challenging for a user, especially when the user deals with hundreds of comments that are scattered around multiple different conversations. A promising solution to this problem is to automatically mine the major topics from conversations and organize them into a hierarchical structure. However, the resultant topic hierarchy can be noisy and/or it may not match the user’s current information needs. To address this problem, we introduce a novel human-in-the-loop approach that allows the user to revise the topic hierarchy based on her feedback. We incorporate this approach within a visual text analytics system that helps users in analyzing and getting insights from conversations by exploring and revising the topic hierarchy. We evaluated the resulting system with real users in a lab-based study. The results from the user study, when compared to its counterpart that does not support interactive revisions of a hierarchical topic model, provide empirical evidence of the potential utility of our system in terms of both performance and subjective measures. Finally, we summarize generalizable lessons for introducing human-in-the-loop computation within a visual text analytics system.

LitPathExplorer: a confidence-based visual text analytics tool for exploring literature-enriched pathway models.

Pathway models are valuable resources that help us understand the various mechanisms underpinning complex biological processes. Their curation is typically carried out through manual inspection of published scientific literature to find information relevant to a model, which is a laborious and knowledge-intensive task. Furthermore, models curated manually cannot be easily updated and maintained with new evidence extracted from the literature without automated support. We have developed LitPathExplorer, a visual text analytics tool that integrates advanced text mining, semi-supervised learning and interactive visualization, to facilitate the exploration and analysis of pathway models using statements (i.e. events) extracted automatically from the literature and organized according to levels of confidence. LitPathExplorer supports pathway modellers and curators alike by: (i) extracting events from the literature that corroborate existing models with evidence; (ii) discovering new events which can update models; and (iii) providing a confidence value for each event that is automatically computed based on linguistic features and article metadata. Our evaluation of event extraction showed a precision of 89% and a recall of 71%. Evaluation of our confidence measure, when used for ranking sampled events, showed an average precision ranging between 61 and 73%, which can be improved to 95% when the user is involved in the semi-supervised learning process. Qualitative evaluation using pair analytics based on the feedback of three domain experts confirmed the utility of our tool within the context of pathway model exploration. LitPathExplorer is available at http://nactem.ac.uk/LitPathExplorer_BI/. sophia.ananiadou@manchester.ac.uk. Supplementary data are available at Bioinformatics online.

Impact of Prominent Themes in Clinician-Patient Conversations on Caregiver's Perceived Quality of Communication with Paediatric Dental Visits.

Patients’ perceived satisfaction is a key performance index of the quality health care service. Good communication has been found to increase patient’s perceived satisfaction. The purpose of this study was to examine the impact of the prominent themes arising from clinician-patient conversations on the caregiver’s perceived quality of communication during paediatric dental visits. 162 video recordings of clinical dental consultations for 62 cases attending the Paediatric Dentistry Clinic of The Prince Philip Dental Hospital in Hong Kong were captured and transcribed. The patients’ demographic information and the caregiver’s perceived quality of communication with the clinicians were recorded using the 16-item Dental Patient Feedback on Consultation skills questionnaires. Visual text analytics (Leximancer™) indicated five prominent themes ‘disease / treatment’, ‘treatment procedure related instructions’, ‘preparation for examination’, ‘positive reinforcement / reassurance’, and ‘family / social history’ from the clinician-patient conversation of the recorded videos, with 60.2% of the total variance in concept words in this study explained through principal components analysis. Significant variation in perceived quality of communication was noted in five variables regarding the prominent theme ‘Positive reinforcement / reassurance’: ‘number of related words’ (p = 0.002), ‘number of related utterances’ (p = 0.001), ‘percentage of the related words in total number of words’ (p = 0.005), ‘percentage of the related utterances in total number of utterances’ (p = 0.035) and ‘percentage of time spent in total time duration’ (p = 0.023). Clinicians were perceived to be more patient-centered and empathetic if a larger proportion of their conversation showed positive reinforcement and reassurance via using related key words. Care-giver’s involvement, such as clinicians’ mention of the parent, was also seen as critical to perceptions of quality clinical experience. The study reveals the potential of the application of visual text analytics software in clinical consultations with implications for professional development regarding clinicians’ communication skills for improving patients’ clinical experiences and treatment satisfaction.

The Ideological Import of a Second Identity: A Critical Image Analysis to Global EFL Materials

The critical analysis of images or visual texts, as an emerging and adapted version of critical discourse analysis (CDA), can contribute to distinguish bias or fair gender manifestation. The rationale behind this study is that little systematic and exhaustive studies have been conducted with an eye towards investigating images solely in EFL textbooks. To offset this imbalance, this paper adopted Fairclough’s (2001) critical discourse analysis (CDA) and social semiotic analysis (SSA) framework by Kress and van Leeuwen (2006) to interrogate the extent and types of gender bias in the Four Corners series, the commercially-produced English language textbooks (Richards and Bohlke, 2012). By adopting a quantitative and qualitative design, image typologies were tallied, girded and tabulated to see how the probable and hidden portrayal image patterns had been rendered for both genders. The results revealed over-representation persists in all the textbook series in certain degrees for both genders and while a textbook series might over-represent one gender it does not necessarily imply that the individual textbook over-represents the same gender.

Topic- and Time-Oriented Visual Text Analysis.

To facilitate the process of converting textual data into actionable knowledge, visual text analysis has become a popular topic with active research efforts contributed by researchers worldwide. Here the authors present the benefits of combing text analysis (topic models in particular) with interactive visualization. They then highlight examples from prior work on topic- and time-oriented visual text analysis and discuss challenges that warrant additional future research.

Visual Text Analysis in Digital Humanities

AbstractIn 2005, Franco Moretti introduced Distant Reading to analyse entire literary text collections. This was a rather revolutionary idea compared to the traditional Close Reading, which focuses on the thorough interpretation of an individual work. Both reading techniques are the prior means of Visual Text Analysis. We present an overview of the research conducted since 2005 on supporting text analysis tasks with close and distant reading visualizations in the digital humanities. Therefore, we classify the observed papers according to a taxonomy of text analysis tasks, categorize applied close and distant reading techniques to support the investigation of these tasks and illustrate approaches that combine both reading techniques in order to provide a multi‐faceted view of the textual data. In addition, we take a look at the used text sources and at the typical data transformation steps required for the proposed visualizations. Finally, we summarize collaboration experiences when developing visualizations for close and distant reading, and we give an outlook on future challenges in that research area.

Visual analytics for text-based railway incident reports

The GB railways collect about 150,000 text-based records each year on potentially dangerous events and the numbers are on the increase in the Close Call System. The huge volume of text requires considerable human effort to its interpretation. This work focuses on visual text analysis techniques of Close Call records to extract safety lessons more quickly and efficiently. This paper treats basic steps for visual text analysis based on an evaluation test using a pre-constructed test set of 150 Close Call records for “Trespass”, “Slip/Trip hazards on site” and “Level crossing”. The results demonstrate that visual text analysis can be used to identify the risks in a small-scale test set but differences in language use by different cohorts of people interferes with straightforward risk identification in larger sets. This work paves the way to machine-assisted interpretation of text-based safety records which can speed up risk identification in a large corpus of text. It also demonstrates how new possibilities open up to develop interactive visualisations tools that allow data analysts to use text analysis techniques for risk analysis.

Analysis of Rural and Urban Supply Chains in the European School Fruit Scheme Using Qualitative Interviews and Network Analysis of Content

ABSTRACTThis study analyzes the determinants of the positive and negative effects experienced by suppliers in the context of programs involving the joint participation of the state, society, and local economy. The European Union’s School Fruit Scheme (SFS) forms the basis of this study. Particular attention is paid to the origin of the suppliers; that is, whether they are based in rural or urban areas. We interviewed 28 suppliers participating in the SFS in the German Federal State of North Rhine Westphalia. Visual text analytics using network analysis of content was applied using classic forms of qualitative coding and a weighted c-index. Our findings reveal that the negative effects of the scheme are greater in rural areas. Therefore, changes in the framework may be necessary to strengthen rural areas. Further, the social relations are often more pronounced in rural areas compared to those in urban areas.

Machine learning meets visualization for extracting insights from text data

Historically, text mining methods have been enriched substantially by both statistical learning and symbolic AI. Different approaches have been extensively applied over the last 30 years to extract "knowledge" from text. However, in scenarios where the path from data to decisions is unclear, or where different users may be interested in different solutions, the involvement of the user or analyst in the text mining process becomes crucial. Visual Text Analytics aims at addressing these problems by incorporating concepts from Visual Analytics to text mining and natural language processing (Keim et. al, 2010).

THE VICISSITUDES OF THE HUMANITIES

The aims of the article: – to track the change of approaches to humanitarian knowledge; – to define possible ways of its development at the present stage. Methods. The methodology of the research is based on the content analysis of verbal and visual texts, and inductive techniques allowing us to generalize heterogeneous approaches. Results and scientific novelty. The author, having postulated a double objective to trace, on the one hand, how the attitude to humanities changed during centuries, and, on the other, to draft possible routes of further development, scrutinizes the concept homo duplex starting from the early renaissance period, when humanities were singled out of the common knowledge about the human being, when the component of meaning «down to earth, not divine» shaped the kernel of the field of knowledge, later named as humanities. This type of knowledge was opposed to the theology with its concept of the divine origin of man. The author attracts Leonardo da Vinci drawing to define the major research methodology of humanities. Further tackling the evolution of the concept from the homo duplex to homo triplex, the author proves that the dynamics of econceptualization owed to a great extend to the possibility to extrapolate methods of sciences onto humanities research. Touching upon the reason of vicissitudes and recent crisis in humanities the author points to the «doctrinal way of thinking» caused on the one hand by political situation and on the other hand by the «fine art» of turning humanities postulates into a dogma. To prove the point, the author attracts verbal and visual texts, as well as historical facts. Upon this basis the contemporary structure of the humanitarian knowledge is being scrutinized that rests upon the integration of humanities and neurosciences, the research kernel including brain, mind, language and subconsciousness. The author constructs the model of heterogeneous spheres, including biosphere, noosphere, technosphere, semiosphere, socisphere, and ethnosphere. Hence, a new paradigm of research approaches that rests upon «the anthropocentric trinity»: thinking (processing) individual+ self and world creating individual + an individual who communicates both with nature and computer. Practical significance. The proposed model of heterogeneous spheres enables to outline intersections variants of sciences and humanities thus defining new conceptual meaning of homo triplex.

Using visual text analytics to examine broadcast interviewing

Broadcast interview hosts are increasingly adopting hybrid forms of interview through the utilization of interview techniques from different genres within the one interview (Ekström and Kroon Lundell, 2011; Montgomery, 2008). Methods that can visually represent interviews in their entirety have the potential to assist in tracking and tracing genre shifts within a single interview. In this paper we examine traditional genres and hybrid forms of broadcast interviewing using a visual text analytic software Discursis (Angus et al., 2013, 2012a, 2012b). Discursis provides visual representations of whole interviews at-a-glance as well as the ability to focus into particular sections for closer analysis. Drawing on a corpus of 101 interviews from a single television program, this study examines if Discursis can meaningfully visually represent forms of interviewing genres (Montgomery, 2008) and highlight where shifting techniques (Ekström and Kroon Lundell, 2011) are used within a single interview.

Exploratory Visual Analysis and Interactive Pattern Extraction from Semi-Structured Data

Semi-structured documents are a common type of data containing free text in natural language (unstructured data) as well as additional information about the document, or meta-data, typically following a schema or controlled vocabulary (structured data). Simultaneous analysis of unstructured and structured data enables the discovery of hidden relationships that cannot be identified from either of these sources when analyzed independently of each other. In this work, we present a visual text analytics tool for semi-structured documents (ViTA-SSD), that aims to support the user in the exploration and finding of insightful patterns in a visual and interactive manner in a semi-structured collection of documents. It achieves this goal by presenting to the user a set of coordinated visualizations that allows the linking of the metadata with interactively generated clusters of documents in such a way that relevant patterns can be easily spotted. The system contains two novel approaches in its back end: a feature-learning method to learn a compact representation of the corpus and a fast-clustering approach that has been redesigned to allow user supervision. These novel contributions make it possible for the user to interact with a large and dynamic document collection and to perform several text analytical tasks more efficiently. Finally, we present two use cases that illustrate the suitability of the system for in-depth interactive exploration of semi-structured document collections, two user studies, and results of several evaluations of our text-mining components.

As contribuições do gênero textual anúncio publicitário para o ensino- aprendizagem do francês: a capacidade de argumentar e a variedade linguística e cultural através da linguagem verbal e visual

Este artigo objetiva mostrar como a produção de anúncios publicitários pode contribuir para: i) a exposição dos alunos à pluralidade tanto linguística quanto cultural que raramente encontramos em livros didáticos; ii) o desenvolvimento da capacidade de argumentação dos alunos, por meio dos recursos verbais e visuais que caracterizam esse gênero textual. Para tanto, mostraremos o estudo do ensino-aprendizagem de língua estrangeira através de uma sequência didática e da produção de textos pertencentes ao gênero textual anúncio publicitário por alunos de Francês como Língua Estrangeira (FLE). Daremos especial atenção à linguagem verbal e visual articuladas para o ensino do francês e às contribuições do gênero anúncio publicitário para a compreensão dos aspectos culturais e linguísticos veiculados nos textos. Para isso, apoiamo-nos no quadro teórico-metodológico do interacionismo sociodiscursivo (BRONCKART, 1999, 2003, 2006, 2008), no que se refere à compreensão do agir humano através da atividade de linguagem e na união das categorias da Gramática do Design Visual (KRESS e VAN LEEUWEN, 2006), para nos auxiliar na análise de textos visuais. Palavras-chaves: Gênero textual anúncio publicitário. Interacionismo sociodiscursivo. Gramática do Design Visual. Ensino-aprendizagem. FLE. Capacidade de argumentação.

How Hierarchical Topics Evolve in Large Text Corpora.

Using a sequence of topic trees to organize documents is a popular way to represent hierarchical and evolving topics in text corpora. However, following evolving topics in the context of topic trees remains difficult for users. To address this issue, we present an interactive visual text analysis approach to allow users to progressively explore and analyze the complex evolutionary patterns of hierarchical topics. The key idea behind our approach is to exploit a tree cut to approximate each tree and allow users to interactively modify the tree cuts based on their interests. In particular, we propose an incremental evolutionary tree cut algorithm with the goal of balancing 1) the fitness of each tree cut and the smoothness between adjacent tree cuts; 2) the historical and new information related to user interests. A time-based visualization is designed to illustrate the evolving topics over time. To preserve the mental map, we develop a stable layout algorithm. As a result, our approach can quickly guide users to progressively gain profound insights into evolving hierarchical topics. We evaluate the effectiveness of the proposed method on Amazon's Mechanical Turk and real-world news data. The results show that users are able to successfully analyze evolving topics in text data.

Visualising Cultures: The “European Picture Book Collection” Moves “Down Under”

The potential for picture books in national collections to act as mirrors reflecting the reader’s cultural identity, is widely accepted. This paper shows that the books in a New Zealand Picture Book Collection can also become windows into unfamiliar worlds for non-New Zealand readers, giving them the opportunity to learn more about a context in which indigenous Māori mythology is interwoven with an imported European culture. The stories in the Collection provide many chances for examining visual similarities and differences between cultures and developing intercultural awareness, as well as enhancing visual literacy through the analysis of illustrations. Making reference to the first European Picture Book Collection, which was created in 1996, the authors discuss these issues from a theoretical perspective and explore the practical implications through visual text analysis of several books in the New Zealand Picture Book Collection.

UTOPIAN: User-Driven Topic Modeling Based on Interactive Nonnegative Matrix Factorization

Topic modeling has been widely used for analyzing text document collections. Recently, there have been significant advancements in various topic modeling techniques, particularly in the form of probabilistic graphical modeling. State-of-the-art techniques such as Latent Dirichlet Allocation (LDA) have been successfully applied in visual text analytics. However, most of the widely-used methods based on probabilistic modeling have drawbacks in terms of consistency from multiple runs and empirical convergence. Furthermore, due to the complicatedness in the formulation and the algorithm, LDA cannot easily incorporate various types of user feedback. To tackle this problem, we propose a reliable and flexible visual analytics system for topic modeling called UTOPIAN (User-driven Topic modeling based on Interactive Nonnegative Matrix Factorization). Centered around its semi-supervised formulation, UTOPIAN enables users to interact with the topic modeling method and steer the result in a user-driven manner. We demonstrate the capability of UTOPIAN via several usage scenarios with real-world document corpuses such as InfoVis/VAST paper data set and product review data sets.

Semantic Network Analysis as a Method for Visual Text Analytics

This paper proposes an approach on a method for visual text analytics to support knowledge building, analytical reasoning and explorative analysis. For this purpose we use semantic network models that are automatically retrieved from unstructured text data using a parametric k-next-neighborhood model. Semantic networks are analyzed with methods of network analysis to gain quantitative and qualitative insights. Quantitative metrics can support the qualitative analysis and exploration of semantic structures. We discuss theoretical presuppositions regarding the text modeling with semantic networks to provide a basis for subsequent semantic network analysis. By presenting a systematic overview of basic network elements and their qualitative meaning for semantic network analysis, we describe exploration strategies that can support analysts to make sense of a given network. As a proof of concept, we illustrate the proposed method by an exemplary analysis of a wikipedia article using a visual text analytics system that leverages semantic network visualization for exploration and analysis.