Iskanje po katalogu digitalne knjižnice

Opis: The SCP Foundation, an extensive collaborative creative writing project, is internationally recognized for its fictitious scientific articles about securing, containing, and protecting anomalies, also known as SCPs. Since it is an open-source wiki, anyone can submit articles, and the authors are anonymous. This paper focused on researching language and gender in SCP articles. Gender is socially constructed and does not necessarily coincide with biological sex. Gender stereotypes reflect expectations about social groups and their members. They can affect occupational choices and even the health-risk behaviors of men and women. In language, gender stereotypes can be reflected in sexism and slurs. Data for this paper are drawn from content analyses of 50 randomly selected SCP entries. Qualitative data was then summarized into quantitative data. Results indicate a neutral, impersonal, and descriptive language for anomalies, which are referred to with the neuter reference because they are deemed borderline beings. Men are more often positively and negatively regarded than women, as women appear significantly less frequently than men in the entries, regardless of the role examined. Men do, however, appear in the roles of victims more often. Gender-stereotypical professions were associated with men, whereas women were associated with professions similar to those of men, indicating a shift from gender stereotypes. Sexism is reflected in the form of gendered slurs. No other socially constructed genders were found in the SCP entries.
Ključne besede: SCP Foundation, SCPs, language, gender, gender stereotypes, slurs
Objavljeno v DKUM: 15.09.2025; Ogledov: 0; Prenosov: 16
Celotno besedilo (4,93 MB)

Ključne besede: chatbots, artificial intelligence, generative artificial intelligence, large language models, English studies
Objavljeno v DKUM: 27.08.2025; Ogledov: 0; Prenosov: 6
Celotno besedilo (123,95 KB)
Gradivo ima več datotek! Več...

Opis: Intentional bilingualism is becoming increasingly popular in EFL countries, where many second-language speakers of English raise their children as a bilingual, with English alongside the local language. This article explores a case of intentional bilingualism spanning the first four years of a child, Ipek, who lives in Türkiye and has been exposed to Turkish and English since birth. As part of a longitudinal qualitative study, field and reflective notes were analysed to outline the key elements of Ipek's context, including the participants, tools, methods and techniques. The insights and experiences gained in the study may contribute to developing effective strategies for developing intentional bilingualism.
Ključne besede: intentional bilingualism, child bilingualism, family language policy, EFL, Turkish as L1
Objavljeno v DKUM: 29.07.2025; Ogledov: 0; Prenosov: 2
Celotno besedilo (503,93 KB)
Gradivo ima več datotek! Več...

Opis: Background: Domestic violence (DV) is a significant public health concern affecting the physical and mental well-being of numerous women, imposing a substantial health care burden. However, women facing DV often encounter barriers to seeking in-person help due to stigma, shame, and embarrassment. As a result, many survivors of DV turn to online health communities as a safe and anonymous space to share their experiences and seek support. Understanding the information needs of survivors of DV in online health communities through multiclass classification is crucial for providing timely and appropriate support. Objective: The objective was to develop a fine-tuned large language model (LLM) that can provide fast and accurate predictions of the information needs of survivors of DV from their online posts, enabling health care professionals to offer timely and personalized assistance. Methods: We collected 294 posts from Reddit subcommunities focused on DV shared by women aged ≥18 years who self-identified as experiencing intimate partner violence. We identified 8 types of information needs: shelters/DV centers/agencies; legal; childbearing; police; DV report procedure/documentation; safety planning; DV knowledge; and communication. Data augmentation was applied using GPT-3.5 to expand our dataset to 2216 samples by generating 1922 additional posts that imitated the existing data. We adopted a progressive training strategy to fine-tune GPT-3.5 for multiclass text classification using 2032 posts. We trained the model on 1 class at a time, monitoring performance closely. When suboptimal results were observed, we generated additional samples of the misclassified ones to give them more attention. We reserved 184 posts for internal testing and 74 for external validation. Model performance was evaluated using accuracy, recall, precision, and F1 -score, along with CIs for each metric. Results: Using 40 real posts and 144 artificial intelligence–generated posts as the test dataset, our model achieved an F1 -score of 70.49% (95% CI 60.63%-80.35%) for real posts, outperforming the original GPT-3.5 and GPT-4, fine-tuned Llama 2-7B and Llama 3-8B, and long short-term memory. On artificial intelligence–generated posts, our model attained an F1 -score of 84.58% (95% CI 80.38%-88.78%), surpassing all baselines. When tested on an external validation dataset (n=74), the model achieved an F1 -score of 59.67% (95% CI 51.86%-67.49%), outperforming other models. Statistical analysis revealed that our model significantly outperformed the others in F1 -score (P=.047 for real posts; P<.001 for external validation posts). Furthermore, our model was faster, taking 19.108 seconds for predictions versus 1150 seconds for manual assessment. Conclusions: Our fine-tuned LLM can accurately and efficiently extract and identify DV-related information needs through multiclass classification from online posts. In addition, we used LLM-based data augmentation techniques to overcome the limitations of a relatively small and imbalanced dataset. By generating timely and accurate predictions, we can empower health care professionals to provide rapid and suitable assistance to survivors of DV.
Ključne besede: domestic violence, online health communities, large language models, generative artificial intelligence, artificial intelligence
Objavljeno v DKUM: 22.07.2025; Ogledov: 0; Prenosov: 4
Celotno besedilo (780,00 KB)
Gradivo ima več datotek! Več...

Opis: Background Artificial intelligence (AI) is increasingly applied in medicine, with growing interest in its potential to improve outcomes in cardiac arrest (CA). However, the scope and characteristics of current AI applications in resuscitation remain unclear. Methods This scoping review aims to map the existing literature on AI applications in CA and resuscitation and identify research gaps for further investigation. PRISMA-ScR framework and ILCOR guidelines were followed. A systematic literature search across PubMed, EMBASE, and Cochrane identified AI applications in resuscitation. Articles were screened and classified by AI methodology, study design, outcomes, and implementation settings. AI-assisted data extraction was manually validated for accuracy. Results Out of 4046 records, 197 studies met inclusion criteria. Most were retrospective (90%), with only 16 prospective studies and 2 randomised controlled trials. AI was predominantly applied in prediction of CA, rhythm classification, and post-resuscitation outcome prognostication. Machine learning was the most commonly used method (50% of studies), followed by deep learning and, less frequently, natural language processing. Reported performance was generally high, with AUROC values often exceeding 0.85; however, external validation was rare and real-world implementation limited. Conclusions While AI applications in resuscitation demonstrate encouraging performance in prediction and decision support tasks, clear evidence of improved patient outcomes or routine clinical use remains limited. Future research should focus on prospective validation, equity in data sources, explainability, and seamless integration of AI tools into clinical workflows.
Ključne besede: Cardiac arrest, Resuscitation, Artificial intelligence, Machine learning, Deep learning, Large language model, Scoping review
Objavljeno v DKUM: 22.07.2025; Ogledov: 0; Prenosov: 3
Celotno besedilo (1,45 MB)
Gradivo ima več datotek! Več...