Text Analyses Research Articles

Neural natural language processing for long texts: A survey on classification and summarization

The adoption of Deep Neural Networks (DNNs) has greatly benefited Natural Language Processing (NLP) during the past decade. However, the demands of long document analysis are quite different from those of shorter texts, while the ever increasing size of documents uploaded online renders automated understanding of lengthy texts a critical issue. Relevant applications include automated Web mining, legal document review, medical records analysis, financial reports analysis, contract management, environmental impact assessment, news aggregation, etc. Despite the relatively recent development of efficient algorithms for analyzing long documents, practical tools in this field are currently flourishing. This article serves as an entry point into this dynamic domain and aims to achieve two objectives. First of all, it provides an introductory overview of the relevant neural building blocks, serving as a concise tutorial for the field. Secondly, it offers a brief examination of the current state-of-the-art in two key long document analysis tasks: document classification and document summarization. Sentiment analysis for long texts is also covered, since it is typically treated as a particular case of document classification. Consequently, this article presents an introductory exploration of document-level analysis, addressing the primary challenges, concerns, and existing solutions. Finally, it offers a concise definition of “long text/document”, presents an original overarching taxonomy of common deep neural methods for long document analysis and lists publicly available annotated datasets that can facilitate further research in this area.

AUTOMATED TEXT ANALYSIS OF HISTORICAL DOCUMENTS USING MACHINE LEARNING TECHNIQUES

This paper presents an innovative approach to the automated analysis of historical documents through the application of advanced machine learning techniques. The problem background in automated text analysis of historical documents using machine learning techniques is efficiently processing large volumes of diverse historical texts to extract valuable insights and patterns, enhancing historical research and understanding. Leveraging the power of natural language processing, this study proposes a comprehensive framework that encompasses data preprocessing, feature extraction, and model training, enabling the efficient extraction of valuable insights from vast collections of historical texts. By utilizing cutting-edge algorithms, including deep learning and sentiment analysis, the research demonstrates the potential of this approach in uncovering hidden patterns, sentiments, and semantic nuances within historical documents, thereby facilitating a deeper understanding of the past and shedding light on critical historical events and societal developments.

An Ecocritical Exploration of Bakhtin's Major Concepts in Solar Storms

Mikhail Bakhtin is a Russian scholar who has been studied internationally in fields of literature and the humanities. Basing his theory on the principle of communication, Bakhtin formed a significant path in literary and cultural studies. Some scholars like Michael Gardiner (1993), Michael M. Bell (1994), Michael J. McDowell (1996), Timo Müller (2010) and Patrick D. Murphy (2013) propose that Bakhtin’s concepts can be appropriated to ecocriticism to make more powerful analyses of literary texts since both Bakhtinian literary criticism and ecocriticism highlight diversity, heterogeneity, agency, and interaction. In this way, Bakhtinian ecocritical theory suggests a new definition of the human subject in its relation to the physical environment and nonhuman beings. Revealing the viability of Bakhtinian critical theory with ecological concerns, this article aims to study the novel Solar Storms by Linda Hogan, a Native American female writer, through an “eco-Bakhtinian” approach to explore how nature is represented in Native American context, how the relationship between human and nonhuman worlds is depicted, how both worlds function in one another, how and why the perception of nature differs in cultural aspects, and how the concept of nature has changed over time. In doing so, this article attempts to show the interaction between culture-nature, human-nonhuman, earth-body, body-mind, traditional-modern, native-non-native, fact-fiction, story-history, self-other, individual-society, and text-reader.

Automatic analysis of public health service text based on character level convolutional neural network

The text information processing technology of public health service is one of the hot research topics at present. To improve the defects of public health service texts, such as inaccurate word segmentation, spelling errors and professional vocabulary understanding, this study designed a character-level deep neural network model on the characteristics of public health service texts. In this model, the bidirectional short and short time memory and the attention pooling operation layer are introduced to make the model better classify the text according to the context. In addition, counter perturbation is introduced in this study to improve the robustness and generalization ability of the model, thus improving its classification effect. The performance verification results show that the proposed model has better classification performance on the public health service text data set. The anti-disturbance samples generated by the model are all in the range of 0–0.2 when WMD deviation degree is measured, while most of the other methods are in the range of 0.4–0.6. The experimental object of this study is ultrasonic examination data. The experimental results show that the automatic analysis model of public health service text based on character level convolutional neural network constructed in this study has excellent accuracy and convergence speed, and has excellent performance in the classification of public health service text in different subject areas.

Capitalizing on the Moment: The Strategic Role of Information Disclosure in Online Crowdfunding

Online crowdfunding platforms have created new avenues for investors to finance existing businesses or new business ventures of project creators. Project creators who launch crowdfunding campaigns can provide progress updates to potential investors throughout the fundraising process, helping to alleviate potential investors’ uncertainty about the project’s success. However, limited attention has been given to the strategic decisions of project creators on when to provide campaign updates. In this study, we focus on the timing strategy of information disclosure and investigate the effectiveness of campaign updates at different timing and performance level. Using a dataset from a leading online crowdfunding platform, we find that updates are more effective at the earlier stage of a crowdfunding campaign in attracting investors, and they play a different role depending on the project’s fundraising performance. For underperforming campaigns, the effectiveness of updates is more salient at the early stage of the campaign, whereas we find no differential impact at the later stage. By employing text analyses on the project descriptions and updates, we also derive insights about how the content of updates impacts their effectiveness at different stages of the campaign. Finally, our results also show that the timing of updates is more important for less experienced project creators, whereas the content of updates is more important for experienced project creators. These results provide important insights for project creators to strategically manage their campaigns in online crowdfunding platforms.

Speaking of civilians: Automated text analysis of the United Nations’ framing of complex humanitarian emergencies

Since the end of the Cold War, the protection of civilians has increased its weight on the United Nations (UN) agenda. This article (1) maps the evolution of civilian protection within the UN framework as an indicator of its shifting priorities, (2) identifies breakpoints in the prevalence and character of the protection discourse, and (3) explores how internal processes of policy development and real-world triggers (namely conflicts and peacekeeping operations) shaped this transformation. The article uses Structural Topic Modeling (STM) to analyze an original corpus of Security Council and General Assembly resolutions on complex humanitarian emergencies since 1990. The analysis uncovers two distinct forms of protection, labeled as “Ground protection” and “Political–legal protection,” which are characterized by contrasting temporal and geographic trajectories. Moreover, critical junctures in the protection rhetoric (during the years 2000, 2005, and 2008) coincide with policy watersheds rather than conflict outbreaks or trends in peacekeeping deployment. This article offers a comprehensive analysis of the intricate evolution of civilian protection using text-as-data methods, which uses its theory-building design to encourage further explorations on the interplay between internal and external factors in shaping its progression within the UN framework.

‘I was right here, the whole time, none of you could see me’: Background Ghosts, Fear, and Vision in Mike Flanagan’s The Haunting (2018–2020)

This article focuses on Mike Flanagan’s Haunting series and brings the pervasive background ghost to the foreground of analysis by studying that which is, as expressed by young Nell in The Haunting of Hill House, ‘right here the whole time’ and yet cannot be seen. Using Jacques Derrida’s seminal notion of ‘spectre’ and its related ‘visor effect’ concept to build textual analyses of both The Haunting of Hill House and The Haunting of Bly Manor, this study demonstrates the ways in which the background ghost both reflects and challenges conventional horror dynamics of visuality and spectacularity. Such dynamics extend beyond the story space and involve the implied and near-sighted spectator, at once imperfect ghost-hunter and hunted by the background ghost. Such pervasive mechanics of haunting allow Flanagan to create a space of highly contemporary terror oozing from the screen to the implied viewer.

Games as Cognitive Recreation: User Perspectives on Brain-Training Apps

This article investigates the role of games in cognitive training through multimethod deductions from user references to the phenomenon, a consideration absent in the current literature. The results of qualitative thematic and quantitative textual analyses on a corpus of commercial brain-training apps’ reviews (N = 30,000) paint a complicated picture of user perceptions of brain-training services, where recreation from gaming takes precedence over concerns of cognitive gains. While users remain uncertain about benefits in cognitive functions, positive perceptions suggest the affective utility of playing brain-training games and the self-regulatory benefits of using the apps. These findings suggest a link between brain-training and psychological variables associated with complex entertainment experiences from playing digital games. The critical role of entertainment from gameplay and associated psychological dimensions are discussed as they relate to our findings and emerging perspectives on the role of gaming in cognitive and affective functions.

Formation of Some Elements of the Right to Defence in Misdemeanour Proceedings

Abstract The text analyses normative solutions concerning the right to defence in Polish proceedings in cases of offences. It discusses the formation of selected elements of the right to defence, from the pre-war solutions to modern times. The author highlights the autonomy of the solutions in force in the 2001 Code of Proceedings in Misdemeanour Cases, and their similarities to the right to defence in criminal cases, constituting the most comprehensive model of this right.

Large-scale text analysis using generative language models: A case study in discovering public value expressions in AI patents

Abstract We put forward a novel approach using a generative language model (GPT-4) to produce labels and rationales for large-scale text analysis. The approach is used to discover public value expressions in patents. Using text (5.4 million sentences) for 154,934 US AI patent documents from the United States Patent and Trademark Office (USPTO), we design a semi-automated, human-supervised framework for identifying and labeling public value expressions in these sentences. A GPT-4 prompt is developed that includes definitions, guidelines, examples, and rationales for text classification. We evaluate the labels and rationales produced by GPT-4 using BLEU scores and topic modeling, finding that they are accurate, diverse, and faithful. GPT-4 achieved an advanced recognition of public value expressions from our framework, which it also uses to discover unseen public value expressions. The GPT-produced labels are used to train BERT-based classifiers and predict sentences on the entire database, achieving high F1 scores for the 3-class (0.85) and 2-class classification (0.91) tasks. We discuss the implications of our approach for conducting large-scale text analyses with complex and abstract concepts. With careful framework design and interactive human oversight, we suggest that generative language models can offer significant assistance in producing labels and rationales.

Trzy tancerki flamenco – trzy transkulturowe historie. Tożsamość transkulturowa dziesięć lat później

This text analyses the explanatory potential of the concept of transcultural identity, inscribed in the broader framework of terms such as “transculturality” and “transculturation.” The author refers to a study of foreign flamenco dancers in Seville she conducted ten years ago. In the new research, she goes back to three of her respondents to learn about their subsequent transcultural steps. New, in-depth interviews are confronted with the previous research and analysed through the lens of transcultural identity extended by the concept of transcultural capital.

CONNECTING THE DOTS THROUGH INTERDISCIPLINARY LITERARY STUDIES: PSYCHOANALYSIS, COSMOPOLITANS, MENTAL HEALTH, POVERTY, AND URBANISM. BOOK REVIEW: ANKHI MUKHERJEE’S UNSEEN CITY: THE PSYCHIC LIVES OF THE URBAN POOR, 2021, OXFORD UNIVERSITY PRESS. 263 PAGES.

This paper reviews the conceptual and theoretical framework propounded by Ankhi Mukherjee. Her framework is developed through rigorous analyses of literary texts from diverse genres. It is philosophically underpinned in psychoanalysis and postcolonial studies. The interdisciplinary framework is eclectic in nature and it can be used in trauma studies, ageing population, mental health, health communication, and urban studies. This paper delineates the development of the framework and its application.
 
 Keywords: Psychoanalysis, postcolonialism, interdisciplinary analysis, population studies, urbanism.
 
 Cite as: Chong, S. T., Abd Rahman, A. Z., & Abu Bakar, R. (2024). Connecting the dots through interdisciplinary literary studies: Psychoanalysis, cosmopolitans, mental health, poverty, and urbanism. (Book review). Journal of Nusantara Studies, 9(1), 139-142. http://dx.doi.org/10.24200/jonus.vol9iss1pp139-142

Developing an empathic analysis: Using critical literacy, dialogue, and inquiry with literature to explore the issues with gender labels

AbstractBook banning and censorship in the U.S. prompts necessary conversations on how critical literacy, dialogue, and inquiry are used in various school and library settings. We share guiding questions alongside three examples of textual analyses centering on gender fluidity with three young adult novels. We believe that English language arts teachers will benefit from seeing examples of how we responded to these texts with critical analysis questions that require students to analyze gender representation and identity within the stories. By developing an empathic analysis, students can engage in this critical work within classrooms and libraries, where reading diverse texts is encouraged.

The gender representation of women and men in the occupational areas of STEM and care work in German textbooks

Science, technology, engineering, and mathematics (STEM) occupations and care work occupations are highly segregated by gender. School textbooks play an essential socializing role in determining which occupations are perceived as typically male or female. Existing research on the gender representation of STEM and care work occupations in textbooks is limited in scope. Therefore, we used quantitative text analyses in a large sample of 202 current German textbooks to examine the gender representation of STEM and care work occupations. We used collocation analysis to explore the nature of the occupational representations, focusing on agency and communion. Men were portrayed significantly more frequently than women in STEM and care work occupations. Adjectives of agency and communion occurred rarely in the collocations. Further research is required to test our findings in other cultures and to take a more differentiated look into the use of agency and communion in textbooks.

Political In Between: Streaming Stand-Up Comedy and Feminist Reckoning in Contemporary Mainland China

China’s strict censorship not only depoliticizes stand-up comedy but creates obstacles for feminist activities. With an experimental approach that compares comedy shows online and offline, this paper asks in which ways comedy and feminism can gesture the political critique and resistance around public culture. Based on seven-month participant observation in comedy clubs in mainland China, I analyze comments and criticism toward women stand-up comedians and offer textual analyses of comedians’ performances in a popular online variety show, Roast & Rock. Beginning with an issue that a woman professional comedian’s performances received laughter and applause offline but criticisms online, I reveal the twisted joking and laughing relationships in women’s stand-up comedy and in turn argue that the debates over “what is the proper women’s voice in public culture,” or what I call feminist reckoning, have become politicized feminist expression that is performed and embodied in-between online streaming and live club comedies. Revealing the (re)politicization of stand-up comedy, I envision possibilities that make feminist media practices visible in mainland China’s public culture.

Discourses of risk in public health advertising about underage alcohol consumption

Abstract Underage alcohol consumption has been linked to major risks, the practice of which has been linked to parental responsibility and management and a major topic in public health advertising campaigns. The aim of the current study is to investigate how risk discourses are constructed in public health advertisements against underage alcohol consumption. To achieve this aim, I apply the framework of social semiotics to analyse a series of public health campaigns in Australia to understand the construction of risk across multiple meaning-making resources. The analyses of the advertisements provide insights into the choices of semiotic resources to construct multiple perspectives of risk discourses to communicate alcohol-related risks and persuade parents to not supply alcohol to their children. The study opens up new agendas of research, by complementing textual analyses with audience interpretations, to provide deeper insights into how alcohol-related public health advertising is perceived, as well as improve its effectiveness and impact in the real world.

Reasoning in classroom dilemma situations: how pre-service teachers judge performance assessment

This study examines pre-service teachers’ reasoning structures based on their beliefs in the context of school performance assessment. We used reflective writing to investigate pre-service teachers’ judgment and reasoning regarding different functions of performance assessment. Forty-five undergraduate pre-service teachers participated in our study. Using a mixed-method approach, we conducted categorial and reconstructive text analyses as well as exploratory statistical analyses to describe the participants’ reasoning structures. Such cognitive structures comprise potential solutions to the performance assessment dilemmas that teachers face in their everyday teaching practice. We found varying distributions of and relationships between (individual-, objective-, social-, and ability-related) reference norms (neutral, student-, and teacher-centered) reference perspectives as well as (causal-analytic, normative, descriptive, and effect-oriented) modes of argumentation. Our discussions related to future research activities on teachers’ reasoning structures in the classroom.

Discourse Networks of the Far Right: How Far-Right Actors Become Mainstream in Public Debates

ABSTRACT How do far-right actors and issues structure public debates and become mainstream over time? Previous research has shown that far-right actors are central actors in political conflicts in Western Europe, partly due to their dominance in struggles on cultural issues. The entrenchment of the far right substantially depends on their networks and public visibility. However, we still lack a long-term understanding of public debates about far-right actors and their discursive interconnections across issues. In this paper, we deliver the first longitudinal analysis of the evolution and structure of far-right discourse networks and the diffusion of far-right ideas in public debates since the 1990s. Methodologically, this paper innovates by combining discourse network analysis with automated text analysis to investigate how the relations between far-right actors and issues are represented in mass media. We analyze more than 500,000 newspaper articles in six different mass media outlets in Germany. The results show that far-right actors have gained more public visibility since the 1990s. The study demonstrates that the mass media debates have introduced and amplified cultural issues “owned” by the far right. Far-right discourse networks in mass media have grown, discursive interactions have intensified and discourse networks have become more centralized. Our results indicate that the discursive mainstreaming of the far right evolved in the context of the “refugee crisis” in 2015, and was linked to the AfD’s institutional access. Overall, the study aids in developing a better understanding of the success of the far right and the crucial role of public debates.

Students Error Analysis in Descriptive Text of UNCP

This research deals with the the kinds of errors of the first semester students of Elementary School Teacher Education Study Program of Palopo Cokroaminoto University mostly do in their writing. The population of this research is the first semester students of Elementary School Teacher Education Study Program of Palopo Cokroaminoto University in academic year 2023. The total number of population was 200 students. The researcher applied the purposive sampling technique and there are 20 students as sample. Writing test is the instrument of this research. Based on the findings and discussion in the previous chapter, the researcher concludes the kinds of errors of the first semester students of Elementary School Teacher Education Study Program of Palopo Cokroaminoto University mostly do in their writing is selection. It can be proven by the kinds of error made by the students are selection error (21 errors or 53.84%), omission error (9 errors or 23.07%), and addition error (9 errors or 23.07%).

Improved BTM topic embedding method for Web text data extraction

With the popularity of social software such as Facebook and Twitter, online text data has begun to explode. However, due to the sparseness and imbalance of text data, it is difficult to accurately analyze and process data such as emotions contained in online text. Therefore, according to the characteristics of Web text, the Web text data extraction model based on biterm topic model, bidirectional recurrent neural network, topic embedding and attention mechanism, namely BTM-BRNN-TEAM model, is proposed to improve the accuracy of emotion analysis in short text. The test results show that when the first 10 topic words are selected, the recognition accuracy, error rate, precision, recall rate and F1-measure of this model are about 77.6%, 22.4%, 83.4%, 76.4%, and 0.82, respectively, which are higher than those of the other two models. The area under the receiver operating characteristic curve of the model is about 0.91, which indicates that its performance is superior. In the processing of short text sentences, the proposed model is able to accurately identify aspectual words that have a significant impact on the text, while weakening the impact of non-aspectual words. The above results show that the Web text processing model proposed in this study has superior performance in sentiment analysis and semantic recognition.