https://www.theseus.fi:443/handle/10024/702897 2026-04-23T17:28:46Z https://www.theseus.fi:443/handle/10024/912450 Assessing Hybrid Financial Distress Prediction Models Using Fusion Methods: Evidence from Finnish Listed Firms Al Nabulsi, Nasib This thesis evaluates a hybrid approach for forecasting financial stress among Finnish listed companies using quarterly data. The framework combines (i) numerical financial indicators modeled with logistic regression, Extreme Gradient Boosting (XGBoost), and a multilayer perceptron (ANN), and (ii) Management Discussion and Analysis (MD&A) disclosures encoded using FinBERT sentence embeddings and aggregated with a TextCNN feature extractor. Two fusion strategies are assessed: feature-level concatenation and weighted ensembling of predicted probabilities. Using 50 repeated experiments under two training sampling settings (1:1 and 1:3), numerical models provide strong performance at the 3-quarter horizon, with logistic regression yielding stable ranking and separation (AUC/KS). In text-only comparisons, a FinBERT baseline is competitive at 3Q, while the contribution of richer text features becomes more pronounced at the 5-quarter horizon, where FinBERT-TextCNN performs more favorably. Across settings, the preferred fusion strategy depends on both the classifier and the evaluation criterion, reflecting trade-offs between ranking performance (AUC), maximum separability (KS), and fixed-threshold accuracy (ACC). Consistent with prior evidence that direct concatenation can introduce redundancy, weighted ensemble fusion more often yields the strongest overall results, particularly at the longer horizon. Overall, the results indicate that quarterly distress forecasting benefits from combining modalities, with numerical indicators dominating short-horizon predictability and textual disclosures providing complementary value as the forecast horizon lengthens. 2026-03-16T06:09:41Z https://www.theseus.fi:443/handle/10024/910489 Operationalizing KYC in Finland: Banks’ Obligations and Practices under National, EU and International Anti-Money Laundering/Counter Terrorism Financing Frameworks Koivula, Nora This thesis examines how Finnish banks prevent money laundering (ML) and terrorist financing (TF), focusing on legal obligations and institutional responsibilities within Finland’s financial-crime prevention frameworks. The aim is to find out how Finnish law assigns and defines a banks’ obligations in AML/CTF, how international institutions recommend banks to proceed and how these obligations cascade to the banks and especially to their KYC operations. The research is limited to the banking sector and applies a qualitative content analysis method to examine systematically regulatory texts and authoritative guidance, supported by theoretical lenses drawn from risk-based compliance, regulatory governance and financial crime prevention frameworks. The material consists of Finnish legislation and EU directives, supervisory guidelines from the Finnish Financial Supervisory Authority (FIN-FSA) and other relevant institutional literature on financial crime prevention. The analysis finds that banks in Finland face extensive obligations derived from several regulatory levels and that they play a critical role in preventing ML/TF in the banking system cascading to a bank’s complicated and multi-step Know Your Customer (KYC) processes. The EU law and the Finnish AML/CTF Act set strict rules for banks to follow when it comes to due diligence, monitoring, reporting and internal controls. The bank’s responsibilities include risk assessment, ongoing customer monitoring, escalations of suspicious transactions and maintenance of governance structures capable of ensuring effective ML/TF risk management. In the end the Finnish banks face a complex and evolving regulatory environment, where effective ML/TF prevention relies on the integration of legal compliance, collaborative work in order to perform necessary KYC processes, and organizational capability. The on-going regulatory development will determine how these duties are interpreted and implemented in practice. 2026-02-03T10:18:59Z https://www.theseus.fi:443/handle/10024/909405 Adoption of Artificial Intelligence to Support Healthcare and Clinical Education: A Scoping Review on Policymakers' and Users’ Perceptions in Nigeria and South Africa. Onyemaucheya, Kelechi Recent advancements in healthcare enhance personalised care and preventive medicine, and they also support advanced clinical trials. Healthcare technologies transform the traditional linear approach to a nonlinear one, enabling faster access to practice and informed decision-making. Artificial intelligence (AI) in healthcare integrates advanced technologies into healthcare systems to promote efficient, personalised medicine; precision interventions targeting specific diseases; enhanced healthcare accessibility; predictive analytics; and drug discovery. AI in healthcare has demonstrated significant benefits in reducing healthcare disparities and promoting equitable access to quality care. AI systems encompass machine learning, deep learning, natural language processing, and computer vision. AI systems process vast amounts of data to recognise patterns, understand and respond to language, make predictions, and interpret images using algorithms. This research aims to highlight the need for AI adoption in clinical practice and to elucidate its uptake among policymakers and users perceptions in both countries. My primary focus will be on Nigeria and South Africa, given their impact on healthcare in Africa. The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA-ScR) protocol for scoping review is employed in a multi-approach data collection and analysis. Relevant scientific databases were searched for articles published from 2020 to 2025 on the following topics: “Artificial intelligence”, “Healthcare technologies”, “Healthcare policies & AI”, “Policymakers in Nigeria and South Africa”,” AI adoption” and “Health professionals ”. AI in clinical practice within healthcare frameworks will support clinicians in preventive interventions for early disease detection and treatment planning. It will educate them on integrating these AIs into personalised care without compromising ethics or privacy. Identified gaps include vague financial constraints, inadequate training, and the infrastructure required to overhaul the traditional method of personalised care. These gaps need to be addressed by future research so that healthcare policymakers can develop effective frameworks for governing Artificial Intelligence. Keywords: Artificial Intelligence, healthcare policymakers, Nigeria, South Africa, Healthcare policies and frameworks. 2026-01-07T08:22:51Z https://www.theseus.fi:443/handle/10024/909391 Motivationsfaktorer i babysimarbete hos anställda babysiminstruktörer vid Folkhälsan: en kvalitativ fallstudie Seunavaara, Vivian Detta examensarbete har gjorts för Folkhälsan. Syftet med studien är att kartlägga vilka faktorer som påverkar motivationen hos fastanställda babysimningsinstruktörer som arbetar på Folkhälsan. Motsvarande tidigare studier har inte gjorts. I arbetet användes metodtriangulering: en elektronisk webbenkät och en temaintervju. Med hjälp av metoden kunde både kvalitativa och kvantitativa data samlas in. Vid analysen av materialet användes kategorisering. Den teoretiska referensramen utgörs av självbestämmandeteorin, som försöker förklara människans motivation genom tre psykologiska grundläggande behov: autonomi, kompetens och samhörighet. Svarens innehåll i den elektroniska enkäten var likartat, och med hjälp av temaintervjuerna fördjupades det tidigare insamlade materialet. Som resultat av studien framkom olika huvudkategorier som både främjar och försvagar babysimningsinstruktörernas motivation. Faktorer som främjar motivationen är delaktighet, autonomi, kompetens och samhörighet. Faktorer som försvagar motivationen är bristande ledarskapsstöd, ojämlikhet, avbrytande av verksamheten samt tidsbrist. Slutsatsen är att babysimningsinstruktörernas motivation huvudsakligen styrs av inre motivationsfaktorer och att självbestämmandeteorins psykologiska grundbehov uppfylls. De faktorer som i arbetet framkommer som försvagande för motivationen kan utmana trivseln i arbetet. Som fortsättningsrekommendation föreslås att ledningen uppmärksammar dessa faktorer och att studien, på grund av det låga antalet respondenter, genomförs på nytt med målet att öka svarsmängden. 2026-01-07T07:27:13Z