Corpus Data Research Articles

Dominant Language, Urbanization and Lexical Depletion of Native Language: A Corpus Linguistics Study

India is home to hundreds of languages. Each Indian language is unique and an identity marker for individuals, communities, and their culture. The English language, though not native to India, has a significant presence in this vast country. One hundred and ninety years of British colonial rule over India had established English as the ‘dominant language’ in the country long ago. Rapid urbanization; and the consequent proliferation of telecommunication networks have further increased its dominance over Indian languages in the last few decades. The heavy code-mixing of English words in the Indian languages is reducing the richness of the vocabulary of the Indian languages by replacing their words with English words, thereby initiating ‘lexical depletion’ based language change in the native languages. This paper is an empirical study to map the code-mixing of English words in the Hindi language and its impact on the Hindi language, as observed in Greater Noida, a town that has seen rapid urbanization since 1991. This paper attempts to explain the implications of code-mixing on the Hindi lexicon by analysing a corpus of spoken language data using corpus-linguistics and computational linguistics methodologies.

The syntactic position of the “people” in the political discourse of left and right populism (NLP-based syntactic analysis experience)

This paper attempts to measure populism in English-language speeches of politicians using computational linguistics methods. The relevance of this study is related not only to the rise of populism in the world and the importance of understanding the mechanisms of political discourse, but also to the lack of linguistic research in the context of corpus studies. Most of the methods proposed to date require significant resources or suffer from structural limitations, especially when they rely heavily on the human annotation process to analyze extensive corpus data. The material for the study was public speeches and interviews of right-wing populist politician and 45th President of the United States Donald Trump and political rhetoric of left-wing populist politician Senator Bernie Sanders. The corpus analysis was based on transcripts downloaded from The American Presidency Project and Miller Center websites. The syntactic position of key lexical units is determined using a comprehensive method based on dependency grammar combined with natural language processing (NLP) and van Leeuwen’s linguistic theory. The application of natural language processing methods confirms the assumption that the frequency of politicians’ reference to the people is not the only relevant indicator for measuring people-centrism in (populist) political discourse, as usually proposed in the political science literature. The results of the study indicate that there is no predominant role of the people in the discourse of Trump and Sanders. In most cases, the people appear in a complementary or adjunct position, indicating their secondary importance in their rhetoric. Given that populism implies proximity to the people and is positioned as “vox populi” – “the voice of the people”, it can be assumed that for both Trump and Sanders, addressing the people is only a way to achieve their political goals.

Telugu language hate speech detection using deep learning transformer models: Corpus generation and evaluation

In today's digital era, social media has become a new tool for communication and sharing information, with the availability of high-speed internet it tends to reach the masses much faster. Lack of regulations and ethics have made advancement in the proliferation of abusive language and hate speech has become a growing concern on social media platforms in the form of posts, replies, and comments towards individuals, groups, religions, and communities. However, the process of classification of hate speech manually on online platforms is cumbersome and impractical due to the excessive amount of data being generated. Therefore, it is crucial to automatically filter online content to identify and eliminate hate speech from social media. Widely spoken resource-rich languages like English have driven the research and achieved the desired result due to the accessibility of large corpora, annotated datasets, and tools. Resource-constrained languages are not able to achieve the benefits of advancement due to a lack of data corpus and annotated datasets. India has diverse languages that change with demographics and languages that have limited data availability and semantic differences. Telugu is one of the low-resource Dravidian languages spoken in the southern part of India.In this paper, we present a monolingual Telugu corpus consisting of tweets posted on Twitter annotated with hate and non-hate labels and experiments to provide a comparison of state-of-the-art fine-tuned deep learning models (mBERT, DistilBERT, IndicBERT, NLLB, Muril, RNN+LSTM, XLM-RoBERTa, and Indic-Bart). Through transfer learning and hyperparameter tuning, the models are compared for their effectiveness in classifying hate speech in Telugu text. The fine-tuned mBERT model outperformed all other fine-tuned models achieving an accuracy of 98.2. The authors also propose a deployment model for social media accounts.

Narrowing or in Kálmán's writings

AbstractThe paper investigates the usage of the Hungarian connective vagy ‘or’. Our starting point is Ariel & Mauri's (2018, 2019) and Ariel's (2020) papers about the use of or, where they argue that its core meaning is ‘alternativity’. Our goal is to describe Hungarian vagy ‘or’ by analyzing various corpus data, and compare the results. We examined the personal subcorpus of the Hungarian National Corpus (MNSZ2), and the Hungarian Spontaneous Speech Database (BEA). In this paper, as a tribute to the memory of László Kálmán, we investigated a third corpus that is constructed from Kálmán's very popular informative texts on Qubit.

Křivda a odpor : konstrukce alternativní modernity a mobilizace amerických konzervativních věřících

This article explores much debated contemporary development in the contestation of the dominant, liberal interpretation of modernity by conservative Christians in the United States through counter hegemonic discourse that is deployed in the political arena and facilitates broader cultural struggle. The examination combines the analysis of discourse from the extensive corpus of data produced by five elite actors in the American religious and political milieu and its interplay with the socioeconomic factors of their constituency on the one hand and the institutional and historical context of accounts. Following in the footsteps of critical sociology of religion, emphasis was placed on processes through which the actors establish a claim for power and legitimize the application of moral rationality as the different path to modern society. Employment of discursive instruments in form of myths, narratives, and negative representations tied to specific religious language, is not only showing the permeability of the church-state separation in supposedly secular political arrangements. More importantly, it is unrevealing the authoritative, mobilizational power of the religious matrix of emotionally motivating meanings and identity-making narratives in face of shortcomings and crises of the hegemonic socio-political constellation. In this manner, it serves as a tool to challenge established interpretations of commitments to freedom and equality, or in many instances reinterprets them to legitimize policy entailed in the notion of collective morality. However, this considerable engagement also creates points of contention in society and in religious formations, which leads to their fragmentation between different interpretations of such commitments.

Understanding Syrian migration in Lebanon: a methodological framework.

Wars, crises, and climate change are just a few of the worldwide concerns that have resulted in the forced relocation of millions of people. After 12 years of conflict in Syria, millions of Syrians are still displaced in the neighbouring countries, and their conditions have worsened due to the economic and socio-political crisis of the region. This paper reports on a study conducted in Lebanon as part of the EU Horizon-funded project ADMIGOV - Advancing Alternative Migration Governance. It describes the methodological framework used to study Syrian migration in Lebanon and sheds light on the phenomenon's patterns, challenges, and impacts. In our study, we opted for a mixed method. It is built on a large corpus of primary data collected over the course of years of intensive, in-depth fieldwork and the author's immersion in the community. Alongside observations, quantitative and qualitative phone interviews were conducted to obtain the perceptions of displaced Syrians living in informal tented settlements in rural Lebanon and an incomplete building in the city of Saida. This interview data is accompanied by primary and secondary data sources, including the findings of other European research projects, statistics from UNHCR and IOM, and academic and press articles. Our research revealed the difficulties Syrians displaced in Lebanon encounter while navigating the challenging situation they are trapped in. Based on a case study approach, it unveils similarities and differences determined by the government's no encampment policy that led to self-settled practices across the country. This approach helped in understanding the challenging dynamic created by weak public institutions and their failure to guarantee the observance of basic human rights, compromising displaced Syrians safety. The weakness of public institutions and their failure to guarantee the observance of basic human rights has compromised displaced safety. Moreover, even though the development interventions and aid assistance have been necessary for Syrians' survival, they proved insufficient, and unequally distributed by location evidencing the inefficiencies of the majority of development aid projects. The findings contribute to an enriched understanding of the situation of Syrians in Lebanon and offer insights for policymakers, practitioners, and researchers working in the field of forced migration and humanitarian responses.

Advancing DNA Language Models through Motif-Oriented Pre-Training with MoDNA

Acquiring meaningful representations of gene expression is essential for the accurate prediction of downstream regulatory tasks, such as identifying promoters and transcription factor binding sites. However, the current dependency on supervised learning, constrained by the limited availability of labeled genomic data, impedes the ability to develop robust predictive models with broad generalization capabilities. In response, recent advancements have pivoted towards the application of self-supervised training for DNA sequence modeling, enabling the adaptation of pre-trained genomic representations to a variety of downstream tasks. Departing from the straightforward application of masked language learning techniques to DNA sequences, approaches such as MoDNA enrich genome language modeling with prior biological knowledge. In this study, we advance DNA language models by utilizing the Motif-oriented DNA (MoDNA) pre-training framework, which is established for self-supervised learning at the pre-training stage and is flexible enough for application across different downstream tasks. MoDNA distinguishes itself by efficiently learning semantic-level genomic representations from an extensive corpus of unlabeled genome data, offering a significant improvement in computational efficiency over previous approaches. The framework is pre-trained on a comprehensive human genome dataset and fine-tuned for targeted downstream tasks. Our enhanced analysis and evaluation in promoter prediction and transcription factor binding site prediction have further validated MoDNA’s exceptional capabilities, emphasizing its contribution to advancements in genomic predictive modeling.

Smuggling in the Grand Duchy of Lithuania–kingdom of Prussia borderland at the end of the eighteenth century: the case of Palanga customs

ABSTRACT Using a surviving set of monthly confiscation reports covering the period between August 1793 and March 1794 (23 cases in total), this article examines smuggling in the territory subordinated to the Palanga customs on the Lithuanian-Prussian border. The article aims to reveal the patterns of smugglers’ behavior, the characteristics of individuals involved in contraband, as well as the state’s response to these illegal economic activities. In addition to other more context-specific conclusions, the research reveals how even a small corpus of data can allow us to analyze the phenomenon of smuggling from a qualitative perspective.

COMPARATIVE ANALYSIS OF SYNTACTIC COMPLEXITY IN INDONESIAN GRADUATE-LEVEL WRITING: A CORPUS STUDY OF MASTER'S THESES VERSUS DOCTORAL DISSERTATIONS

Syntactic complexity is an essential indicator of English Foreign Language (EFL) students' academic writing quality. Previous studies analyzed and compared syntactic complexity in published research articles through large-grained measures. However, the study of syntactic complexity unpublished research at graduate level gains less attentions. Thus, this study tried to fill the gap by analyzing and comparing syntactic complexity of master theses and doctoral dissertations that written by Indonesian students through fine-grained measures. This study conducted a quantitative design of a corpus-based method. The corpus data consists of two sub-corpora that selected through stratified sampling technique based on years and issues. Those were 52 doctoral dissertations and 74 master theses from English Language Education of Universitas Negeri Malang. The twelve phrasal complexity measures were employed through Tool for the Automatic Analysis of Syntactic Sophistication and Complexity (TAASSC) tool created by Kyle (2016). The finding showed the characteristics of Indonesian graduate-level writing that they were intended to utilize more dependents per direct object (standard deviation) measures, and fewer dependents per nominal (standard deviation) measures in their writing. It also showed doctoral dissertations have higher quality than master theses which reflected significant difference in most of phrasal complexity measures. These outcomes offer syntactic complexity insights of Indonesian context. The study highlighted the need more attention to syntactic complexity at graduate-level writing for improving writing quality.

A Hop Away from Everywhere: A View of the Intercontinental Long-haul Infrastructure

We present a longitudinal study of intercontinental long-haul links (LHL) - links with latencies significantly higher than that of all other links in a traceroute path. Our study is motivated by the recognition of these LHLs as a network-layer manifestation of transoceanic undersea cables. We present a methodology and associated processing system for identifying long-haul links in traceroute measurements, and report on our findings from. We apply this system to a large corpus of traceroute data and report on multiple aspects of long haul connectivity including country-level prevalence, routers as international gateways, preferred long-haul destinations, and the evolution of these characteristics over a 7 year period.

Opening up Corpus FinSL: enriching corpus analysis with linguistic ethnography in a study of constructed action

Abstract Corpus methods are now established within the field of signed language linguistics. Empirical investigations of signed language corpora have challenged many early assumptions about the nature of deaf community signed languages, while making us more aware of the limitations of traditional corpus documentation methods. One limitation relates to insufficient documentation of ethnographic information that is often necessary for accurately understanding and interpreting corpus data. Linguistic ethnography offers unique possibilities for addressing this limitation. This article outlines a novel interview method developed to enrich the original Corpus of Finnish Sign Language (Corpus FinSL) with additional ethnographic information eight years after it was first documented and archived with standard IMDI corpus metadata. We interviewed 22 Corpus FinSL signers about their family and social networks, as well as their lifelong language, geographical, educational, and employment trajectories. Here we describe how this information illuminates the linguistic analysis and interpretation of constructed action – an enactment-based way to express meaning – in Corpus FinSL data. Our results reveal constructed action in FinSL discourse is influenced by factors like signer’s educational background and age, but not exclusively by family networks or use of other sign languages. The interview materials demonstrate diversity and change in the communicative ecologies of FinSL, which is discussed in relation to the use of constructed action in FinSL. We argue that this kind of approach enables signed language corpus linguistics to “open up” more to signers’ lived experiences, while still “tying down” empirical descriptions of FinSL. A major benefit is the enrichment of both machine-readable annotations and metadata, while supporting deeper engagement between deaf signing communities and signed language corpus projects.

A modal account of syntactically non-integrated von wegen in contemporary German

Abstract This study aims to account for the German expression von wegen (literally ‘of ways’) when used outside of the sentence structure. Drawing on corpus data for written and spoken German, we show that there are two types of syntactically non-integrated von wegen: (i) illustrating/exemplifying von wegen and, (ii) opposing von wegen. By combining both a formal and a functional perspective, we claim that these two types can be analyzed as modal expressions. In fact, from a formal point of view based on Werner Abraham’s classification in lexical and grammatical forms of modality, both illustrating/exemplifying and opposing von wegen encode a quotative/evidential meaning, whereas only the second type expresses an epistemic modal value. Nevertheless, von wegen does not belong to the prototypical forms of modality described by Abraham as it does not occupy a structural position within the sentence and therefore does not fall into any of the classes proposed by Abraham. For this reason, we also need to integrate the formal perspective with a functional one by defining modality not only structurally but also semantically. In doing so, we account for the modality of von wegen in that we assume the existence of a third strategy of modalization that is specific to syntactically non-integrated expressions and operates at the discourse level.

Public Ceremony: The Construction of Nationalism Through Mobile Film Screening Under Jiang Kai-shek’s CNP Government

The Chinese Nationalist government initiated the first national network of mobile film screenings, which subsequently had a great influence on political propaganda in Chinese cinema. Based on corpus data from Chinese archives, diaries, newspapers, periodicals, and literature archived from 1931 to 1945, this study investigates the goals, effects, and shortcomings of mobile film exhibition under Jiang Kai-shek’s Chinese Nationalist Party (CNP) government in China, focusing on the Second Sino-Japanese War. I contend that the Nationalist government regarded films as propaganda to promote cultural identity, the social state, and collectivist ideology. To some extent, such films contributed to the development of democracy and national morality before the war, persuasion during the war, and consolation after the war. However, the mobile film industry in the Jiang era also suffered from some practical problems, such as the lack of planning, inefficient management, obscure content, and chaotic viewing settings. Hence, mobile projection remained a form of public ceremony whose impact deviated from its original intent.

Jugendsprache: bahasa remaja Jerman pada akun Instagram Funk

German youth language or Jugendsprache is a language style used by German teenagers and young adults in their daily communication. This language style is adopted by a German content network named Funk in its Instagram account posts to gain attention from its audience, which specifically targets people aged between 14 to 29 years old. The purpose of this research is to investigate the forms of Jugendsprache used by Funk in their Instagram posts. The design of this research is descriptive qualitative. The data corpus in total of 288 sentences was acquired from Funk’s Instagram account. The results unveiled that there were 177 lexicons, 55 word formations and 28 semantic shifts found from the data. Anglicism, in total of 112 data, is the most dominant lexicon phenomenon compared to the others. Another significant process are 29 composition data in word formation and 23 extensions of meaning data in semantic shift. From this research, it was found that English is widely used in Funk’s contents, either in the form of words, phrases, sentences in full English or English words adopted into the German morphological system. Funk utilizes English to make their contents interesting for its’ target audience, the young people.

Aspectual reduplication in Sign Language of the Netherlands: reconsidering phonological constraints and aspectual distinctions

Abstract This study investigates the use of predicate reduplication to express aspectual meaning in Sign Language of the Netherlands (NGT). The study focuses on three aspect types that have been found to be encoded by reduplication across sign languages – habitual, continuative, and iterative – and addresses potential phonological restrictions. Naturalistic corpus data and data elicited from six deaf NGT signers were taken into account. The results suggest that (i) predicate reduplication can express all three aspect types, but it is optional; (ii) reduplication expressing habitual and continuative aspect appears to be phonologically constrained; and (iii) such phonological constraints do not apply to iterative reduplication, whose form is different from the other two aspects, in that the reduplication cycles are separated by pauses. Since there is no formal difference between habituals and continuatives in the data, it is suggested that this semantic distinction may not be grammaticalized in the language, and that, possibly, the inflectional system of NGT instead more broadly distinguishes imperfective/perfective viewpoint. While this latter suggestion is in line with findings reported for many spoken languages, the results are different from what has previously been described for NGT as well as for other sign languages. Potential explanations for these differences can be found in both methodological and sociolinguistic factors.

Early experiences of peer support workers starting employment in a mental health charity: a case study

In recent years a plethora of job roles has emerged across the voluntary and community sector (VCS) and public sector that explicitly request lived experience (LE) of mental health challenges. These roles are often situated in the ‘frontline’ workforce providing direct support to people accessing services. This article shares early learning about the experiences of people who have lived experience of mental health challenges employed as paid peer support workers (PSWs) within a mental health charity. The findings are drawn from fieldwork conducted over a three-month period with five recently employed PSWs, conducted as a component of the author’s doctoral study. The data corpus included interviews, fieldwork observation notes from ‘walk the frontline’ (WTF) activities, and the collation of WhatsApp voice notes sent by the PSWs to the researcher. The article presents nine key themes that emerged from the data and categorises these into three areas: Firm up – those that were broadly positive denoting good practice; Fine tune – those that require further refinement; and Focus – those which signal a need for concentrated attention and further exploration. Taking these findings into account, a tentative schematic model is offered which suggests sequential ‘conditions’ to be considered when developing PSW programmes. This has relevance to voluntary sector organisations (VSOs) who are considering recruiting and deploying PSWs to support frontline service delivery.

Resolution Enhancement of SOHO/MDI Magnetograms

Research on the solar magnetic field and its effects on solar dynamo mechanisms and space weather events has benefited from the continual improvements in instrument resolution and measurement frequency. The augmentation and assimilation of historical observational data timelines also play a significant role in understanding the patterns of solar magnetic field variation. Within the realm of astronomical data processing, super-resolution (SR) reconstruction refers to the process of using a substantial corpus of training data to learn the nonlinear mapping between low-resolution (LR) and high-resolution (HR) images, thereby achieving higher-resolution astronomical images. This paper is an application study in high-dimensional nonlinear regression. Deep learning models were employed to perform SR modeling on SOHO/MDI magnetograms and SDO/HMI magnetograms, thus reliably achieving resolution enhancement of full-disk SOHO/MDI magnetograms and enhancing the image resolution to obtain more detailed information. For this study, a data set comprising 9717 pairs of data from 2010 April to 2011 February was used as the training set, 1332 pairs from 2011 March were used as the validation set and 1034 pairs from 2011 April were used as the test set. After data preprocessing, we randomly cropped 128 × 128 sub-images as the LR cases from the full-disk MDI magnetograms, and the corresponding 512 × 512 sub-images as HR ones from the HMI full-disk magnetograms for model training. The tests conducted have shown that the study successfully produced reliable 4× SR reconstruction of full-disk MDI magnetograms. The MESR model’s results (0.911) were highly correlated with the target HMI magnetographs as indicated by the correlation coefficient values. Furthermore, the method achieved the best PSNR, SSIM, MAE and RMSE values, indicating that the MESR model can effectively reconstruct magnetograms.

Vision Verbs Emerge First in English Acquisition but Touch, not Audition, Follows Second

AbstractWords that describe sensory perception give insight into how language mediates human experience, and the acquisition of these words is one way to examine how we learn to categorize and communicate sensation. We examine the differential predictions of the typological prevalence hypothesis and embodiment hypothesis regarding the acquisition of perception verbs. Studies 1 and 2 examine the acquisition trajectories of perception verbs across 12 languages using parent questionnaire responses, while Study 3 examines their relative frequencies in English corpus data. We find the vision verbs see and look are acquired first, consistent with the typological prevalence hypothesis. However, for children at 12–23 months, touch—not audition—verbs take precedence in terms of their age of acquisition, frequency in child‐produced speech, and frequency in child‐directed speech, consistent with the embodiment hypothesis. Later at 24–35 months old, frequency rates are observably different and audition begins to align with what has previously been reported in adult English data. It seems the initial orientation to verbalizing touch over audition in child–caregiver interaction is especially related to the control of physically and socially appropriate behaviors. Taken together, the results indicate children's acquisition of perception verbs arises from the complex interplay of embodiment, language‐specific input, and child‐directed socialization routines.

Emergent Pragmatic Conventions in Spoken ELF Corpus Data: Micro-Diachronic Analysis of Inclusive vs. Exclusive Multilingual Practices

This article examines multilingual practices as an example of emergent pragmatic conventions in three Transient International Groups (TIGs) using spoken English as a lingua franca (ELF) from the Vienna-Oxford International Corpus of English (VOICE). The analysis combines principles of corpus linguistics and conversation analysis by adopting a new approach for the micro-diachronic analysis of spoken interaction. Quantitative and qualitative evidence and micro-diachronic visualizations of VOICE transcripts show how the three groups examined interactively develop group-specific multilingual practices. The analysis reveals that the three groups have different preferences in this respect. While two groups develop inclusive multilingual practices in the course of their interaction, one group shows a tendency to use multilingual practices exclusively, primarily in side sequences. In addition to multilingual use, the presence or absence of metalinguistic discussions about language (and languages) plays a role for creation of shared transcultural territory and the formation of group identity. These processes are indicative of how unacquainted multilingual speakers negotiate and develop pragmatic conventions more generally.

DISCOURSE: KNOWLEDGE, NEWS, AND FAKE INTERTWINED

Discourse has been a focal point for linguists over an extended period. The multidisciplinary character of the term ‘discourse’ has resulted in diverse approaches aiming to define and explore this phenomenon. The study of various facets of discourse significantly enhances its comprehension. For our understanding, discourse encompasses the following pertinent characteristics: it is a communicative process, a dialogue, an interaction between communicators; it is a structure, a system, a distinct way of representing and perceiving the world, a special language and world, an expression and manifestation of national identity. Discourse implies knowledge, as it includes various types of it. Knowledge and news are closely linked. They are processed in discourse and with the help of it. Any explicit or implicit knowledge and news influence discourse production and comprehension. True and fake information can alter existing types of knowledge and create new ones, influence mental processes, and shape new mental models. The aim of the article is to describe discourse in its close connection with its main constituents – knowledge, news and fake. ‘Fake’ becomes a linguistic and discursive matter. It is associated with the concepts of ‘lie’, ‘deception’, and ‘manipulation’. Corpus data about ‘fake’ and ‘fake news’ give a bigger picture of the overall context of their representation, changes and dynamics in the usage. The widespread influence of fake is provided with the help of mass media. Strategic control of knowledge in the interaction process is achieved through the use of knowledge itself, accordingly, it becomes possible to control knowledge by means of deception and fake. Fake news is a genre of news information and news discourse, it is an integrative type of media text. The main goal of fake news is to deceive and manipulate consciousness of a wide audience. Manipulation entails language to subtly influence recipients, encouraging them covertly to undertake particular actions. Fake news can be studied within news discourse, media discourse, television discourse, manipulative discourse, political discourse, discourse of fake, and other types. Thus, knowledge, news and fake presented in discourse and with the help of it allow to create ‘new knowledge’, shape the audience’s understanding of discourse, make and generate mental models, control and restrict access to certain information, manipulate consciousness of the audience, transform traditional mental attitudes associated with the pursuit of truth.