108495 research outputs found
Sort by
NER som ett källidentifieringsverktyg. Erfarenheter av svenska BERT för digital historia 1.25
Get PDFThe paper explores my experiences of working with Named Entity Recognition (NER) in Swedish parliamentary records. As such, it provides a practical account of my methodology in employing the Swedish BERT and its NER functionality in a historical dataset. It also reflects on the relevance of this case to the broader relationship between digital and
traditional intellectual history. The study described used NER to identify the geographical areas and placenames within Swedish parliamentary discourse from 1887 to 1914. Taken together, this list of locations could be used to determine the aggregate frequencies of geographical groupings, in this case predominantly nations. The quantitative findings were subsequently used to navigate the data set and identify the most relevant texts for qualitative, contextual close readings. This paper argues that there are strengths in employing digital tools but maintaining the framework of traditional intellectual history in accordance with ‘digital history 1.25
Exploring Patient Organization Periodicals with the Topic Timelines Text Visualization Method
Get PDFThe text visualization technique Topic Timelines offers a compact visualization to represent the evolution and clustering of topics over time, while also providing direct access to the texts in which these topics appear. In this paper, we describe how Topic Timelines was further developed within the ActDisease project, by adding functionality for generating timelines using different types of topic extraction techniques and connecting the visualization to existing interfaces for the close reading of texts. Additionally, we evaluate how the updated temporal topic overview can support corpus exploration.
The experiments were conducted on a digitalized corpus from the ActDisease project, consisting of patient organization periodicals from the Swedish Diabetes Association, published between 1949 and 1990. Timelines were generated based on topics extracted using sentence transformers clustering and integrated with the ActDisease text database interface - a user interface developed for exploring and reading texts digitalized within the project
Craft as culture, not artefact : an analysis of the role of museums and heritage centres in Scotland in heritage craft safeguarding
Get PDFThis study explores how Scottish museums and heritage centres safeguard heritage crafts through two case studies: Historic Environment Scotland’s Engine Shed and Skills Training Centre in Stirling, and the Shetland Museum & Archives’ partnership with Eve Eunson. Using a mix of qualitative methods, it examines institutional and practitioner collaborations. The findings reveal the complementary value of formal training and community-led initiatives, emphasizing context-specific safeguarding strategies aligning policy and practice. By documenting initiatives following the UK’s CICH ratification, the study offers timely insights into how cultural institutions support and sustain Scotland’s heritage crafts through engagement, best practices, and adaptive approaches.https://www.ester.ee/record=b6019724*es
Fiction as curatorial practice : rethinking Kyrgyz State Historical Museum through Mukai Elebaev's The Long Way
Get PDFThis thesis explores the curatorial potential of exhibiting a literary novel in a historical museum, focusing on the 1916 Uprising in Central Asia. It examines Mukai Elebaev’s The Long Way and the State Historical Museum in Bishkek as narrative texts, using critical museology, narratology, memory studies, and decolonial theory. Through interdisciplinary analysis and critical museum visit frameworks, the thesis proposes a hypothetical audio-based installation integrating the novel into the museum. Curating emerges as a method blending theory and imagination, aiming to introduce narrative, affective, and decolonial strategies into a post-Soviet, didactic museological context.https://www.ester.ee/record=b6019713*es
Investigating Linguistic Abilities of LLMs for Native Language Identification
No full textLarge language models (LLMs) have achieved state-of-the-art results in native language identification (NLI). However, these models often depend on superficial features, such as cultural references and self-disclosed information
in the document, rather than capturing the underlying linguistic structures. In this work, we evaluate the linguistic abilities of opensource LLMs by evaluating their performance in NLI through content-independent features, such as POS n-grams, function words, and punctuation marks, and compare their performance against traditional machine learning approaches. Our experiments reveal that while LLM’s initial performance on structural features (55.2% accuracy) falls significantly below their performance on full text (96.5%), fine-tuning significantly improves their capabilities, enabling state-of-the-art results with strong cross-domain generalization
Euroopa Liidu väärtused Brexiti protsessis: väärtushinnangute lõhe Ühendkuningriigis ja Euroopa Liidu liikmesriikides
No full textSee töö uurib Ühendkuningriigi ja Euroopa Liidu liikmesriikide väärtushinnangutes toimunud nihkeid Brexiti protsessi ajal. Väärtused, mida käsitletakse tulenevad Euroopa Liidu alusdokumentidest. Kasutades European Social Survey rahvaküsitluste andmeid, analüüsitakse väärtushinnangute trende osapoolte vahel. Rahvaküsitluste küsimused analüüsitakse ja teisendatakse väärtuste mõõdetavateks esitusviisideks. Analüüsis kasutatakse mitteparameetrilisi statistilisi teste tõlgendamaks muutusi väärtustes. Tulemustest selgus, et Euroopa Liidu esindatavate väärtuste poolehoid oli mõlemal osapoolel languse trendis kuni Brexiti protsessi lõppemiseni. Brexiti protsessi lõppemine oli korrelatsioonis Ühendkuningriigis toimunud järsu väärtushinnangute poolehoiu suurenemise nihkega. Euroopa Liidu liikmesriikides stabiliseerus väärtuste poolehoiu vähenemise trend Brexiti protsessi lõppemisel. Väärtused, mille prioritiseerimise erinevus osapooltes suurenes vaadeldud perioodil on võrdsus, inimväärikus, solidaarsus, vabadus, demokraatia. Tulemused viitavad prioriteetide nihkele, ja kontekstuaalse raamistiku olulisusele riigi avalikus arvamuses, mis mõjutab baasväärtuste tõlgendust.https://www.ester.ee/record=b5733251*es
Relationships between adherence and persistance to antihypertensive medication use and Big Five personality traits based on data from the Estonian Biobank
Get PDFRaviplaanist kinni pidamine kogu raviperioodi vältel on maksimaalse ravitoime saavutamiseks oluline. Antud uurimistöö eesmärgiks on vaadelda, milline seos on patsiendi isiksusel ravimkäitumisega, uurides täpsemalt seoseid ravimi võtmise järjekindluse, järjepidevuse ja patsiendi isiksuse vahel. Tuginedes varasemale kirjandusele, vaadati Neurootilisuse ja Meelekindluse seoseid järjekindluse ning järjepidevusega. Valim moodustati perioodil 2020-2021 vererõhuravimeid tarvitanud ning isiksuseküsimustiku täitnud Eesti Geenivaramu geenidoonoritest (N = 6698). Tulemused näitavad, et järjekindlusel on oluline seos sellega, kas ravi katkestati küsitud perioodil või mitte. Suure Viisiku isiksuseomadustest leiti järjekindlusel nõrgad seosed Neurootilisusega ja Ekstravertsusega ning järjepidevusel nõrk seos Avatusega kogemusele. Meelekindlusel seos järjekindluse ja järjepidevusega puudus. Tulemustest nähtub, et isiksuse efekti suurused jäävad pigem väikesteks. Isiksusejoonte seosed raviplaani järgimisega on komplekssed ning vajavad põhjalikumat analüüsi erinevate diagnooside ja ravimite näitel
The use of a service sharing platform to reduce the loan burden on businesses
Get PDFhttps://www.ester.ee/record=b573834
Relationship between symptoms of depression and values
Get PDFKäesoleva uuringu eesmärk oli leida depressiooni soodumuse ja põhiväärtuste vahelisi seoseid. Uuriti ka taustategurite seoseid depressiooni sümptomite ning väärtuste esinemisega. Uuringus osales 166 täiskasvanut, kellest 134 (81%) olid naised ja 28 (17%) mehed ning kelle vanus varieerus 19-70 eluaasta vahel, keskmise vanusega 30 (SD = 13,29). Uuringus osalejad täitsid emotsionaalse enesetunde küsimustiku ning Schwartzi 21 küsimusega portreeküsimustiku. Koguvalimist 50%-l esines soodumus depressioonile ning depressioon oli kõrgem madalama sissetuleku ning haridusega inimestel. Samuti korreleerus vanus depressiooni ning kogemusele avatuse ja eneseupitamise väärtustega negatiivselt. Depressiooni soodumusega ning soodumuseta inimestes ei esinenud statistiliselt olulisi erinevusi põhiväärtustes
Rethinking Low-Resource MT: The Surprising Effectiveness of Fine-Tuned Multilingual Models in the LLM Age
Get PDFThis study challenges the current paradigm shift in machine translation, where large language models (LLMs) are gaining prominence over traditional neural machine translation models, with a focus on English-to-Faroese translation. We compare the performance of various models, including fine-tuned multilingual models, LLMs (GPT-SW3, Llama 3.1), and closed-source models (Claude 3.5, GPT-4). Our findings show that a fine-tuned NLLB model outperforms most LLMs, including some larger models, in both automatic and human evaluations. We also demonstrate the effectiveness of using LLM-generated synthetic data for fine-tuning. While closed-source models like Claude 3.5 perform best overall, the competitive performance of smaller, fine-tuned models suggests a more nuanced approach to low-resource machine translation. Our results highlight the potential of specialized multilingual models and the importance of language-specific knowledge. We discuss implications for resource allocation in low-resource settings and suggest future directions for improving low-resource machine translation, including targeted data creation and more comprehensive evaluation methodologies