Specific Language Research Articles

"Words matter": Judicial discourse on domestic violence in China

The discourse about domestic violence focuses on power and control by the abuser among the victim through different means, one of those is the language. Besides the domestic walls within the abuses are perpetrated, a more institutional and usually regulated language on family violence is traceable in the public discourse, specifically in the People’s Court judgments. In cases regarding family matters, especially divorce proceedings, Chinese judges do often use paternalistic and educational language towards the couple. The aim of this paper is to carry out a preliminary research and analysis on the judicial discourse about domestic violence in order to evaluate to what extent, how and whether these paternalistic and educational nuances emerge in the language and statements used by judges in those cases. This research will be carried out firstly by building a corpus of judgments on “domestic violence” issues, and then by comparing data and analysing the specific language used by the authority using digital tools.

Enhancing EFL Education: The Role of Metadiscourse in Language Assessment

This study was motivated by students' low ability to write and communicate effectively in English. This study aims to analyze the effect of metadiscourse on language assessment in the context of EFL (English as a Foreign Language) education. This study is a qualitative research. The subjects involved in the study were students learning English as a foreign language. The data collection method was through a robust literature review to establish basic knowledge of metadiscourse and its significance in language assessment. The review also included important works on metadiscourse language evaluation theory and research investigating the relationship between metadiscourse and language skills proficiency. Using a comparative perspective, the analysis of data from language assessment utilizes pass-linguistic and pass-cultural insights to complement the knowledge of metadiscourse in EFL education. By analyzing how metadiscourse operates in specific languages and cultural contexts, the researchers found widespread patterns in addition to particular language variations in the use of metadiscourse. The results showed that metadiscourse plays an important role in shaping students' language skills ability, discourse competence, and universal communicative effectiveness. In conclusion, integrating metadiscourse in language evaluation provides great hope for improving EFL training and promoting stronger language consequences.

A text classification method combining in-domain pre-training and prompt learning for the steel e-commerce industry

Purpose With the development of Web information systems, steel e-commerce platforms have accumulated a large number of quality objection texts. These texts reflect consumer dissatisfaction with the dimensions, appearance and performance of steel products, providing valuable insights for product improvement and consumer decision-making. Currently, mainstream solutions rely on pre-trained models, but their performance on domain-specific data sets and few-shot data sets is not satisfactory. This paper aims to address these challenges by proposing more effective methods for improving model performance on these specialized data sets. Design/methodology/approach This paper presents a method on the basis of in-domain pre-training, bidirectional encoder representation from Transformers (BERT) and prompt learning. Specifically, a domain-specific unsupervised data set is introduced into the BERT model for in-domain pre-training, enabling the model to better understand specific language patterns in the steel e-commerce industry, enhancing the model’s generalization capability; the incorporation of prompt learning into the BERT model enhances attention to sentence context, improving classification performance on few-shot data sets. Findings Through experimental evaluation, this method demonstrates superior performance on the quality objection data set, achieving a Macro-F1 score of 93.32%. Additionally, ablation experiments further validate the significant advantages of in-domain pre-training and prompt learning in enhancing model performance. Originality/value This study clearly demonstrates the value of the new method in improving the classification of quality objection texts for steel products. The findings of this study offer practical insights for product improvement in the steel industry and provide new directions for future research on few-shot learning and domain-specific models, with potential applications in other fields.

Design of Oral English Teaching Quality Evaluation System Based on AHP-Fuzzy Algorithm

In the evolving landscape of English oral language instruction, the integration of online and offline (blended) teaching methodologies has emerged as a significant approach to enhance learning outcomes. This study addresses the need for a robust and nuanced evaluation framework capable of assessing the quality of blended English oral language teaching. Motivated by the limitations of traditional evaluation methods in capturing the complexities and subjective nuances of blended teaching environments, we propose a novel evaluation model that leverages the Analytic Hierarchy Process (AHP) and Fuzzy Association Analysis. This paper outlines a systematic approach to construct a comprehensive evaluation indicator system tailored to the specific demands of English oral language teaching. Utilizing AHP, we determine the relative weights of various evaluation indicators, laying a solid foundation for a structured assessment framework. Subsequently, Fuzzy Association Analysis is employed to handle the inherent uncertainties and subjectivities associated with teaching quality assessment, thereby facilitating a more accurate and holistic evaluation. An empirical case study is presented to demonstrate the practical application of our proposed AHP-Fuzzy model in evaluating the quality of a specific blended English oral language teaching program. The results underscore the model’s effectiveness in providing a detailed and nuanced assessment of teaching quality, highlighting its superiority over traditional evaluation methods in addressing the complexities of blended teaching environments.

Connecting and Parzelled Constructions Located After The Full Stop as a United Structural-Semantic Formation of the Russian Language

Such means of expressive syntax as connecting and parceled constructions were discussed in the article. Discussion questions were presented. In particular, the issue of the proximity of adjunction and parcellation, the status of adjunct constructions located outside the main sentence after the period, was considered. Using methods of description, observation, comparison, analysis and synthesis, the author offers an original understanding of the essence of connecting and parceled structures. They are still interpreted ambiguously in the linguistic literature. Much attention is paid to the criteria for their delimitation and analysis of the famous postulate of V.V. Babaytseva. Contrary to the opinion of the professor and her supporters, the author is convinced that there is no annexation without dismemberment and, conversely, dismemberment without annexation. These are interdependent processes that cannot exist one without the other. The author believes that the formation of parceled structures takes place in four stages, which are determined by short, intensive procedural actions. In the composition of parceled structures, as research shows, two varieties are clearly visible: a) parceled structures and b) parceled connecting structures. The scientific value of the work lies in the proof of the closeness of connecting and parceled constructions, supported by both the methods of their formation and semantics. The practical significance of the study lies in the possibility of using its results to analyze specific language material, including in the process of teaching Russian language to students at Kazakh universities. The conclusion that connecting and parceled constructions located outside the main sentence, after a period, should be considered as a single structural-semantic formation of the Russian language was made.

Are risk disclosures in financial reports informative? A text mining-based perspective

This paper studies whether the risk factors disclosed in financial reports are as informative as SEC requires. The subprime crisis is used as a typical real-risk event to test their informativeness by text mining methods. By analyzing the textual attributes and specific contents of the risk disclosures in 14089 Form 10-K statements of 1,685 financial firms from 2006 to 2022, the results show that firms tend to disclose risks in more specific and negative language and significantly reduce stickiness and boilerplate during the subprime crisis. This paper also finds that the changes of risk factors are consistent with the actual risk profile before, during and after the subprime crisis. The failed institutions during the crisis have disclosed more risks related to the subprime crisis using a more negative tone than the survived institutions. These findings all provide evidence that the risk factors in Form 10-K are informative and not invariable boilerplate.

Embedding the Calendar and Time Type System in Temporal Type Theory

Temporal Type Theory (TTT) has been recently introduced as a topos-theoretic approach to understanding the behaviour of systems over time. A truly innovative point of TTT is that it makes truth inherently dependent on time; this is to be contrasted with the classical approach in which past, present and future are related via logical operators. Further on this line of research, the notion of truth is substituted by the ‘time duration’ over which a proposition is true, giving rise to the logic of temporal landscapes. In this paper, we embed the main notions of the Calendar and Time Type System (a framework that was built for reasoning about calendric notions in the context of the Web and is centred around the notion of time granularity) in Temporal Type Theory, attempting to address a common requirement in the practical use of temporal formalisms: the ability to speak about certain time moments and appropriate time windows in (past and future) time, in the form of hours, dates, holidays, etc. With a view to concrete applications, we provide examples on how calendar types can be combined with temporal operators of TTT to augment the expressive power of this framework. In that respect, the addition of calendar types genuinely adds to the range of Temporal Type Theory as a powerful specification language.

Using DSLs to manage consistency in long-lived enterprise language specifications

Using DSLs to manage consistency in long-lived enterprise language specifications

Language and Cognitive Impairments in Multiple Sclerosis: a Comparative Study of RRMS and SPMS Patients.

Previous studies have reported that patients with Multiple Sclerosis (MS) face cognitive difficulties. Much less, however, is known about their language abilities. The present study aims to provide a clear view of the language abilities of adults with MS, considering their cognitive skills and the type of the disorder. Specifically, 20 patients with Relapsing-Remitting MS, 17 patients with Secondary Progressive MS, and two healthy control groups demographically matched with the clinical groups, took part in the study. Their language abilities were examined using the Adult Language Abilities Test, which explores morpho-syntactic and semantic phenomena of the Greek language in both comprehension and production modes. Our results did not reveal statistically significant differences between the two clinical groups on any cognitive or language subtest. Both groups, however, performed poorer than the control groups in producing the correct form of verbs, possessive and weak personal pronouns, and in comprehending relative and passive clauses. Participants' performance in morphosyntax was found to be related to their processing speed. Mapping the cognitive and language abilities of patients with MS is crucial for designing effective targeted interventions, which can help them adopt strategies to manage their specific cognitive and language symptoms.

Familia sobre todo, emprendimiento ante todo: Secretaría Nacional de la Familia (SNF/MMFDH) y la articulación entre neoconservadurismo y neoliberalismo

The centrality of a certain notion of family in neoconservatism has a superstructural dimension as a reaction to the demands of the feminist and LGBTQIAP+ movements, and a structural dimension that reveals the way it is related to neoliberalism. From the process of confessionalization of public policies, they promote changes in the traditional patterns of the State-society binomial, in its structures, processes and actors that claim a “subsidiary” ideal of state action compared to other social spheres, especially the family (neoconservative) and the market (neoliberal). What I seek to examine here is the way in which neoliberalism in the National Secretariat for the Family (SNF/MMFDH) in Brazil aligns with neoconservative political and cultural strategies, where through the circulation of the family category, it capillarizes a specific social language with direct repercussions on the social protection system.

Exploring Current Methods and Trends in Text Summarization: A Systematic Mapping Study

This paper presents a systematic mapping study of the current methods and trends in text summarization, a challenging task in natural language processing that aims to condense information from one or multiple documents into a concise and coherent summary. The paper focuses on applying text summarization for the Malay language, which has received less attention than other languages. The paper employs a three-phased quality assessment procedure to filter and analyze 27 peer-reviewed publications from seven prominent digital libraries, covering 2016 to 2024. The paper addresses two research questions: (1) What is the extent of research on text summarization, especially for the Malay language and the education domain? and (2) What are the current methods and approaches employed in text summarization, with a focus on addressing specific problems and language contexts? The paper synthesizes and discusses the findings from the literature review and provides insights and recommendations for future research directions in text summarization. The paper contributes to advancing knowledge and understanding of the state-of-the-art techniques and challenges in text summarization, particularly for the Malay language.

ParaAntiProt provides paratope prediction using antibody and protein language models

Efficiently predicting the paratope holds immense potential for enhancing antibody design, treating cancers and other serious diseases, and advancing personalized medicine. Although traditional methods are highly accurate, they are often time-consuming, labor-intensive, and reliant on 3D structures, restricting their broader use. On the other hand, machine learning-based methods, besides relying on structural data, entail descriptor computation, consideration of diverse physicochemical properties, and feature engineering. Here, we develop a deep learning-assisted prediction method for paratope identification, relying solely on amino acid sequences and being antigen-agnostic. Built on the ProtTrans architecture, and utilizing pre-trained protein and antibody language models, we extract efficient embeddings for predicting paratope. By incorporating positional encoding for Complementarity Determining Regions, our model gains a deeper structural understanding, achieving remarkable performance with a 0.904 ROC AUC, 0.701 F1-score, and 0.585 MCC on benchmark datasets. In addition to yielding accurate antibody paratope predictions, our method exhibits strong performance in predicting nanobody paratope, achieving a ROC AUC of 0.912 and a PR AUC of 0.665 on the nanobody dataset. Notably, our approach outperforms structure-based prediction methods, boasting a PR AUC of 0.731. Various conducted ablation studies, which elaborate on the impact of each part of the model on the prediction task, show that the improvement in prediction performance by applying CDR positional encoding together with CNNs depends on the specific protein and antibody language models used. These results highlight the potential of our method to advance disease understanding and aid in the discovery of new diagnostics and antibody therapies.

Evolution of Linguistic Markers of Agency, Centrality and Content During Metacognitive Therapy for Psychosis: A Pilot Exploratory Study.

Metacognitive Reflection and Insight Therapy (MERIT) is a form of person-centred psychotherapy that promotes recovery-oriented outcomes by targeting metacognitive capacity. Previous research has shown the feasibility and clinical benefits of MERIT. However, it is not clear whether and how the specific processes targeted by MERIT (e.g., improved sense of agency, self-reflectivity, awareness of others, etc.) are manifested in the patient-therapist communications. In this study, we used natural language processing (NLP) to explore objective linguistic markers of change during MERIT for early psychosis. Five patients underwent 24 sessions each of MERIT, which were recorded and transcribed verbatim. The transcripts underwent NLP to determine the patterns of pronoun use and the frequency of semantic categories related to temporality and mental processing. Mixed linear regressions were used to examine how NLP features evolve over the course of psychotherapeutic sessions. Analyses detected shifts in specific language signals over time including (1) changes in patterns of pronoun usage with more active and central first-person plural pronoun (We); (2) transition in temporal focus of speech from past-focus towards present- and future-focus and (3) increased words representing perceptual and cognitive processes. Our findings suggest NLP can objectively quantify meaningful signals consistent with expected subjective changes in MERIT and potentially other psychotherapeutic interventions.

On the Mental Paralysis Theme in “A Little Cloud” from the Perspective of the Literary Stylistics

The theme of mental paralysis is one of the central and significant themes in James Joyce’s renowned collection of short stories Dubliners. It vividly portrays the life condition of Dubliners as being plagued by a lack of faith and trapped in a state of spiritual paralysis. Among these stories, “A Little Cloud” is the first story in Dubliners that portrays the adult stage of the mental paralysis. After closely reading this story, it becomes evident that Joyce strategically employs specific language choices that hold immense significance in effectively conveying the theme of mental paralysis. Therefore, this thesis aims to employ the theoretical framework of literary stylistics to analyze the linguistic deviations employed by Joyce in this story, trying to explore how Joyce expresses the theme of mental paralysis in his story.

Economic Sociology and Management in the spectrum of modern managerial and economic ideas

The article is devoted to the radical transformation of the structure of knowledge in modern economic sociology and managerial theory, to the transition from slender hierarchies of ideas (the top of which consists of a small number of axioms, and the base of which consists of multiple specific recommendations) to the postmodern market of concepts. The authors made a comparative analysis of such transformation in economic sociology and managerial theory, revealed general regularities and specific factors that determined the difference between the scientific fields of these science. The latter include: 1) subjects conducting scientific research (in managerial theory, along with scientists, managers, consultants, people somehow connected with business were engaged in such practice), 2) a terminology of sciences (in managerial theory, a specific language for communication between managers and employees was actively created), 3) the use of marketing and PR-technologies to promote ideas. The totality of these factors determined a special “landscape” of the subject areas of economic sociology and managerial theory.The article concludes that despite the existing problems caused by multiparadigmality, the emerging “flexible” format of socio-humanitarian knowledge opens new opportunities for creativity in rapid changes of social reality, obsolescence of classical approaches to its cognition, as well as the realization of the variety of phenomena and processes in the field of economics and management.In such situation, each scientific team should choose its own special place in the exhibition of ideas, create its own “exhibits” attractive to the scientific community and the general public. Among the variety of topical and demanded research directions, the article identifies three key thematic blocks reflecting the scientific priorities of the Department of Economic Sociology and Management of Moscow University: 1) institutional changes and institutional context of socio-economic and managerial processes, 2) aging population and digitalization as key factors of economic and management transformation, 3) socio-economic processes and phenomena in the mirror of digital communication.

From words to wealth: Decoding the language markers and compositions of shop names in Banda Aceh

This qualitative study examines the names of 100 local and independent businesses to study the linguistic diversity and features of the linguistic landscape in Banda Aceh, Indonesia. The names here refer to those written in specific languages used in Aceh (Indonesian, Acehnese, English, Arabic, and other indigenous languages spoken in the province). The data was collected from five main roads, capturing the shop names for analysis. The results show that English is the dominant language choice, composing a majority of 45% of signage, showing the influence of globalization and the desire for international appeal. While English appears to convey prestige, some businesses blend it with Indonesian or Acehnese, connecting with local culture and engaging diverse audiences. This deliberate language choice reflects a commitment to cultural preservation and authentic customer experiences. Meanwhile, diverse linguistic combinations, emphasizing global influences, cultural preservation, and strategic branding, were also revealed. The most prevalent linguistic structure of signs is product+X (57%), which highlights product types, followed by place+X (14%) which emphasizes regional or cultural identity. In addition, a name/person’s name+X establishes a personalized connection (6%). Finally, in language markers, shop names reveal globalization’s prevalence and demonstrate efforts to convey a global image. The revitalization marker reflects the commitment to the Acehnese language and cultural pride. Identity markers, which include place, name, abbreviation, product, plant name, and brand, establish trust, familiarity, and resonance, creating a dynamic linguistic and cultural landscape where businesses steer global influences and local heritage to connect authentically with customers. To conclude, shop names in Banda Aceh illustrate a vibrant linguistic and cultural environment where businesses use global influences and local heritage to create meaningful bonds with customers.

PEak: A Single Source of Truth for Hardware Design and Verification

Domain-specific languages for hardware can significantly enhance designer productivity, but sometimes at the cost of ease of verification. On the other hand, ISA specification languages are too static to be used during early stage design space exploration. We present PEak, an open-source hardware design and specification language, which aims to improve both design productivity and verification capability. PEak does this by providing a single source of truth for functional models, formal specifications, and RTL. PEak has been used in several academic projects, and PEak-generated RTL has been included in three fabricated hardware accelerators. In these projects, the formal capabilities of PEak were crucial for enabling both novel design space exploration techniques and automated compiler synthesis.

Formalizing Sustainable Urban Mobility Management: An Innovative Approach with Digital Twin and Integrated Modeling

Background: Urban mobility management faces growing challenges that require the analysis and optimization of sustainable solutions. Digital twins (DTs) have emerged as innovative tools for this assessment, but their implementation requires standardized procedures and languages; Methods: As part of a broader methodology for continuous DT validation, this study focuses on the conceptual validation phase, presenting a conceptualization approach through formalization using Specification and Description Language (SDL), agnostic to simulation tools. The conceptual validation was achieved through stakeholder engagement in the Bolzano context, producing 41 SDL diagrams that define both elements common to different urban realities and specific local data collection procedures; Results: The feasibility of implementing this stakeholder-validated conceptualization was demonstrated using Simulation of Urban MObility (SUMO) for traffic simulation and optimization criteria calculation, and its framework SUMO Activity GenerAtion (SAGA) for generating an Activity-Based Modeling (ABM) mobility demand that can be improved through real sensor data; Conclusions: The SDL approach, through its graphical representation (SDL/GR), enables conceptual validation by enhancing stakeholder communication while defining a framework that, while adapting to the monitoring specificities of different urban realities, maintains a common and rigorous structure, independent of the chosen implementation tools and programming languages.

NIMG-63. LEVERAGING LLMS FOR ACCURATE DIFFERENTIATION OF RADIATION NECROSIS AND TUMOR PROGRESSION IN BRAIN MRI REPORTS: A STUDY ON AUTOMATED SCORING AND CLINICAL IMPLICATIONS

Abstract BACKGROUND Accurate differentiation between radiation necrosis and tumor progression or recurrence in patients treated with stereotactic radiosurgery for brain metastases is critical for guiding clinical management. This study leverages the advanced natural language processing capabilities of Meta Llama3, an artificial intelligence (AI) large language model (LLM), combined with prompt engineering, to rapidly categorize brain magnetic resonance imaging (MRI) radiology reports. Our objective was to develop an automated scoring system to classify concern for radiation necrosis, tumor progression, equivocal findings, or stable exams. METHODS Using a comprehensive dataset of reports annotated by expert radiologists, we ran inference on a 70-billion parameter Llama3 model (temperature 0.2, top_p 0.9) with specific prompts designed to capture the nuanced language and diagnostic criteria related to radiation necrosis or tumor progression. RESULTS The first pass was performed on a training dataset of 107 reports and did not predefine the clinical conditions. This demonstrated 43.4% accuracy in scoring when compared to a human user’s classification of each report. Agreement between the human reader and Llama3 was assessed using the Gwet agreement coefficient, AC1=0.411 (99%CI 0.396-0.426). Multiple iterations of targeted prompt engineering were then employed to narrow the definition of radiation necrosis and tumor progression, with specific examples and nuanced language used to achieve a higher degree of accuracy at 72.0%, AC1=0.719 (99%CI 0.717-0.722). DISCUSSION/FUTURE DIRECTIONS This surpasses recent demonstration of lower human-LLM agreement in radiographic score assignment, with further room for calibration on a dataset of several thousand reports. Next, we will correlate automated interpretations with actual clinical management decisions for radiation necrosis (e.g., initiation of steroids, bevacizumab, Laser Interstitial Thermal Therapy, and/or repeat imaging). This automated scoring system holds significant potential for LLMs in clinical applications. Future work will focus on integrating this model into clinical workflows and expanding its capabilities to include longitudinal monitoring of patient outcomes.

Cancer Clinical Trial Participation Amongst Culturally and Linguistically Diverse Patients in Australia.

Clinical trials play a large role in oncological. Many barriers to participation in cancer clinical trials exist, including a patient's status as "culturally and linguistically diverse (CALD)". Globally, it is thought that CALD patients experience lower rates of trial participation, although very few studies quantify rates of cancer clinical trial participation in this group. Our study therefore aims to characterize CALD participation in cancer clinical trials in an Australian setting. We conducted a retrospective analysis of data from the Cancer Clinical Trials Centre at Austin Health, a large tertiary metropolitan cancer center in Melbourne, Australia. Participation in cancer clinical trials between groups was compared using simple descriptive analysis, Chi-squared analysis, and logistic regression. Of 2568 patients (mean age 57.9 years, 37% female) offered entry into a cancer clinical trial between 2018 and 2023, 26% were from a CALD background (n = 678), and 9% had a preferred language other than English (n = 219). A greater proportion of non-CALD patients participated in a cancer clinical trial compared with CALD patients (37% versus 33% respectively, p =0.04). In logistic regression models, Arabic (OR,0.46; 95% CI,0.21-0.93 [univariate]) and Greek (OR,0.54; 95% CI,0.31-0.91 [multivariate]) language groups, along with overall CALD status (OR,0.81; 95% CI,0.67-0.99 [univariate]) were associated with lower cancer clinical trial participation. We found that CALD patients, those born in non-English speaking countries, and specific language groups, were associated with lower cancer clinical trial participation.