108495 research outputs found
Sort by
Political discourses of the pandemic in the United States: a comparative study of the causal stories of COVID-19 in Alabama and Iowa
No full textThe COVID-19 pandemic presented unprecedented challenges globally, necessitating quick policy responses from governments worldwide. This thesis examines the role of political discourse in forming and justifying policies during crises, focusing on the early first-wave responses of the U.S. states of Alabama and Iowa. Using Deborah Stone's (1989) causal stories framework, the study identifies four causal story types: intentional, accidental, inadvertent, and mechanical. Through qualitative content analysis of public communications by Governors Kay Ivey and Kim Reynolds, this thesis explores the relationship between crisis communication and policy stringency, highlighting divergences between theoretical expectations and empirical findings. Contrary to expectation, the intentional causal story was overwhelmingly dominant across both states, irrespective of policy stringency. This suggests that leaders may employ intentional narratives universally in crises to convey a sense of control and action. Also unexpectedly, the accidental causal story appeared more frequently in Alabama, a state with higher policy stringency, than in Iowa. The inadvertent causal story, while limited to Governor Ivey’s rhetoric, emerged earlier than initially anticipated. Lastly, while mechanical causal stories were deemed unlikely to appear in the theoretical framework, they nonetheless appeared in Alabama through religious invocations, adding an unforeseen dimension to the analysis. The results of the analysis contribute to the existing literature through the framework of how political leaders use certain narratives to legitimize policy decisions. Focusing on regional executives fills a gap in the literature left by the predominant study of national executives. These insights deepen the understanding of crisis communication by executives while highlighting the potential for future research to investigate broader contexts, mixed-methods approaches, and comparative analyses across political systems and party affiliations. The study’s limitations in scope reveal possible avenues for future research into the dynamics of crisis communication, testing the universality of intentional causal stories indicated in this study and the conditions under which accidental, mechanical, or inadvertent causal stories might emerge
Comparison of emotional facial expression recognition between humans and FaceReader machine learning software
Get PDFAntud töö eesmärgiks oli uurida, kui täpselt suudab masinõppe tarkvara FaceReader tuvastada Ekmani baasemotsioone ning võrrelda selle tuvastamistäpsust inimhindajate tulemustega. Uuringus osales 144 inimhindajat, kellest olid mehed 19,4% ja 79,9% naised. Inimhindajatele ja FaceReaderile esitati 1-sekundilisi videoklippe, mis kujutasid rõõmu, hirmu ning vastikust, ning eesmärk oli stiimulmaterjali hinnata kuue baasemotsiooni (rõõm, hirm, vastikus, üllatus, viha ja kurbus) põhjal. Tulemused näitasid, et inimesed vastasid õigesti 54,4% juhtudest, samas kui masinõppe täpsus oli 37,4%. Korduvmõõtmiste ANOVA analüüs näitas, et inimhindajad suutsid edukalt ära tunda rõõmu ja vastikust, kuid mitte hirmu. FaceReader tuvastas edukalt rõõmu, kuid vastikust ja hirmu mitte. Sõltumatute rühmade ANOVA tulemuste kohaselt olid inimhindajad FaceReaderist statistiliselt olulisel määral täpsemad hirmu ja vastikuse, kuid mitte rõõmu tuvastamisel. Käesoleva töö tulemused on kooskõlas varasemate uurimustega, viidates, et mittevalideeritud ning dünaamilise stiimulmaterjali hindamisel on inimesed masinõpetest sageli täpsemad
Socioeconomic status and language development in Estonian preschoolers with typical development and developmental language disorder
No full textKäesoleva uurimistöö eesmärgiks on emakeelena eesti keelt kõnelevate eakohase arenguga (EKA) ja arengulise keelepuudega (AKP) laste SMS võrdlemine ning AKP laste sotsiaalmajandusliku staatuse (SMS) ning kõnetesti skooride vaheliste seoste uurimine. Uuringus osalesid eelkooliealised lapsed (21 AKP ja 25 EKA). Perekondade SMS näitajatest vaadeldi läbivalt ema ja isa haridustaset ning pere igakuist keskmist netosissetulekut. Laste keelelise arengu hindamiseks viidi läbi 5-6-aastastele lastele mõeldud standardiseeritud kõnetest.
Kõigi SMS näitajate puhul leiti AKP ja EKA gruppide vahel statistiliselt oluline erinevus: ema ja isa haridustaseme näitel keskmine erinevus, pere sissetuleku näitel madal kuni keskmine erinevus. Nii vanemate haridustaseme kui ka sissetuleku puhul oli AKP grupi keskmine madalam kui EKA grupi puhul. AKP grupi keeletestide skooride ja SMS vahel täheldati kõigi kolme SMS näitaja puhul statistiliselt olulist korrelatsiooni, mis oli ema haridustaseme puhul mõõdukas ning isa haridustaseme ja igakuise netosissetuleku puhul nõrk kuni mõõdukas
Second language Korean Universal Dependency treebank v1.2: Focus on Data Augmentation and Annotation Scheme Refinement
No full textWe expand the second language (L2) Korean Universal Dependencies (UD) treebank with 5,454 manually annotated sentences. The annotation guidelines are also revised to better align with the UD framework. Using this enhanced treebank, we fine-tune three Korean language models—Stanza, spaCy, and Trankit—and evaluate their performance on in-domain and out-of-domain L2-Korean datasets. The results show that fine-tuning significantly improves their performance across various metrics, thus highlighting the importance of using well-tailored L2 datasets for fine-tuning first-language-based, general-purpose language models for the morphosyntactic analysis of L2 data
Annotating Attitude in Swedish Political Tweets
No full textThere is a lack of Swedish datasets annotated for emotional and argumentative language. This work therefore presents an annotation procedure and a dataset of Swedish political tweets. The tweets are annotated for positive and negative attitude. Challenges with this type of annotation is identified and described. The evaluation shows that the annotators do not agree on where to annotate spans, but that they agree on labels. This is demonstrated with a new implementation of the agreement coefficient Krippendorff's unitized alpha
The Danish Idiom Dataset: A collection of 1000 Danish idioms and fixed expressions
No full textInterpreting idiomatic expressions is a challenging task for learners and LLMs alike, as their meanings cannot be deduced directly from their individual components and often reflect nuances that are specific to the language in question. This makes idiom interpretation an ideal task for assessing the linguistic proficiency of large language models (LLMs). In order to test how LLMs handle this task, we introduce a new dataset comprising 1000 Danish idiomatic expressions sourced from the Danish Dictionary DDO (ordnet.dk/ddo). The dataset has been made publicly available at sprogteknologi.dk. For each expression, the dataset includes a correct dictionary definition, a literal false definition, a figurative false definition, and a random false definition. In the paper, we also present three experiments that demonstrate diverse applications of the dataset and aim to evaluate how well LLMs are able to identify the correct meanings of idiomatic expressions
Digimaksete tulevik: suundumused ja häired
Get PDFThe evolution of digital payments have transformed the global financial ecosystem by providing faster, more secure, and more accessible ways to transact. However, despite the widespread adoption of digital payment systems, gaps remain in understanding the drivers of their evolution, the challenges that hinder their potential, and the implications of emerging trends such as blockchain, AI, and IoT. The rapid pace of technological advancement and regulatory changes underscore the need for a comprehensive overview of digital payment systems. This thesis addresses these gaps by exploring the evolution of digital payment systems, their lifecycle, and the various tools that drive this transformation. Using a systematic literature review, it addresses research questions related to the digital payment types, key drivers, benefits, challenges, and future trajectories of digital payments. The study highlights the role of technological advancements such as blockchain, AI, and IoT and the impact of regulatory frameworks in shaping payment innovations. Emerging trends, including cryptocurrencies, central bank digital currencies (CBDCs), and mobile payments, underscore the shift toward decentralization and financial inclusion. Despite significant progress, barriers such as interoperability, cybersecurity, and regulatory complexity persist. This thesis offers an understanding of the current landscape and future directions of digital payments, providing valuable insights for stakeholders, including consumers, businesses, financial institutions, and policymakers.Digitaalsete maksete areng on muutnud ülemaailmset finantsökosüsteemi, pakkudes tehingute tegemiseks kiiremaid, turvalisemaid ja juurdepääsetavamaid viise. Vaatamata digitaalsete maksesüsteemide laialdasele kasutuselevõtule on siiski lünki nende evolutsiooni põhjuste, nende potentsiaali takistavate väljakutsete ja esilekerkivate suundumuste, nagu plokiahel, AI ja asjade internet, mõistmisel. Tehnoloogilise arengu kiire tempo ja regulatiivsed muudatused rõhutavad vajadust tervikliku ülevaate järele digitaalsetest maksesüsteemidest. See lõputöö käsitleb neid lünki, uurides digitaalsete maksesüsteemide arengut, nende elutsüklit ja erinevaid tööriistu, mis seda ümberkujundamist juhivad. Kasutades süstemaatilist kirjanduse ülevaadet, käsitletakse selles uurimisküsimusi, mis on seotud digitaalsete maksete tüüpide, peamiste tegurite, eeliste, väljakutsete ja digitaalsete maksete tulevaste trajektooridega. Uuring tõstab esile selliste tehnoloogiliste edusammude rolli nagu plokiahel, tehisintellekt ja asjade internet ning regulatiivsete raamistike mõju makseuuenduste kujundamisel. Tekkivad suundumused, sealhulgas krüptovaluutad, keskpanga digitaalsed valuutad (CBDC) ja mobiilimaksed, rõhutavad nihet detsentraliseerimise ja finantskaasamise suunas. Vaatamata märkimisväärsetele edusammudele püsivad takistused, nagu koostalitlusvõime, küberjulgeolek ja regulatiivne keerukus. See lõputöö pakub terviklikku arusaama digitaalsete maksete praegusest maastikust ja tulevikusuundadest, pakkudes väärtuslikku teavet sidusrühmadele, sealhulgas tarbijatele, ettevõtetele, finantsasutustele ja poliitikakujundajatele
Recommendations for Overcoming Linguistic Barriers in Healthcare: Challenges and Innovations in NLP for Haitian Creole
No full textHaitian Creole, spoken by millions in Haiti and its diaspora, remains underrepresented in Natural Language Processing (NLP) research, limiting the availability of effective translation tools. In Miami, a significant Haitian Creole-speaking population faces healthcare disparities exacerbated by language barriers. Existing translation systems fail to address key challenges such as linguistic variation within the Creole language, frequent code-switching, and the lack of standardized medical terminology. This work proposes a structured methodology for the development of an AI-assisted translation and interpretation tool tailored for patient-provider communication in a medical setting. To achieve this, we propose a hybrid NLP approach that integrates fine-tuned Large Language Models (LLMs) with traditional machine translation methods. This combination ensures accurate, context-sensitive translation that adapts to both formal medical discourse and conversational registers while maintaining linguistic consistency. Additionally, we discuss data collection strategies, annotation challenges, and evaluation metrics necessary for building an ethically designed, scalable NLP system. By addressing these issues, this research provides a foundation for improving healthcare accessibility and linguistic equity for Haitian Creole speakers
The role of religious beliefs in shaping public attitudes toward EU integration in Serbia
No full textThis thesis examines how religious affiliation, particularly to the Serbian Orthodox Church (SOC), influences public attitudes toward European Union membership in Serbia. The relatively low public support for EU integration in Serbia highlights the importance of identifying contributing factors. Religion emerges as a key determinant, especially in contexts where religious institutions have a significant impact on societal values and political preferences. The Serbian Orthodox Church, beyond its role as a religious authority, serves as a symbol that is deeply tied to Serbian national identity. The thesis uses data from the European Social Survey (ESS10) and applies quantitative methods, including logistic regression, interaction, and variance analysis, to assess the impact of religious affiliation on attitudes toward EU membership, with national attachment as a conditional effect. The findings reveal that being a Serb Orthodox alone does not significantly affect attitudes toward the EU, while religiosity plays a more important role, with higher levels of religiosity associated with stronger opposition to EU membership. Incorporating European attachment as an additional moderator provides a deeper and statistically significant interpretation of the conditional effect of national attachment; as European identity strengthens, the conditional effect of national attachment diminishes among Serb Orthodox individuals. Variability in attitudes was more pronounced among Serb Orthodox individuals, particularly those with higher levels of religiosity, reflecting greater ambivalence on the issue. Overall, the thesis highlights the critical role of religious faith and identity dynamics in shaping public opinion on EU membership in Serbia, offering valuable implications for policymakers aiming to bridge societal divides and strengthen support for EU integration.https://www.ester.ee/record=b5734004*es