Göteborgs universitets publikationer - e-publicering och e-arkiv
Not a member yet
78804 research outputs found
Sort by
The Impact of Regional Challenges on Qatar’s Response in Foreign Policy from 1995 to 2024
No full textThis study explores the impact of MENA regional challenges (the Arab Spring, regime changes, failed states and civil war, and regional crises and conflicts) on Qatar’s foreign policy (QFP) (actions, policies/discourse, and alliances) from 1995 to 2024. Utilizing multiple qualitative approaches (case study, content analysis, process tracing) as research tools of this study. An integrated theoretical framework composed of international relations (IR) theories of realism/neorealism, constructivism, omni-balancing, and small state theories used as a theoretical framework to this study. A tringle method was used to ensure the validity and reliability of the research tools of this study. Findings showed that Qatar's foreign policy (actions, policies, and alliances) has been shaped by regional challenges (the Arab Spring, regime change, failed state, regional crises, and conflicts). In specific, during the Arab Spring, Qatar adopted a proactive foreign policy, focusing on military engagement and support for revolutions. With regard to regime changes, Qatar's actions became more ideologically defined (patron of Islamic-oriented post revolutionary regimes), selective, and quickly engaged with new governments. Moreover, the emergence of failed states and civil wars, Qatar responded with a hyperactive strategy that includes early military engagement, later-stage mediation, long-term diplomacy, media,
humanitarian aid, backchannel alliances, and selective neutrality. Finally, regional crises and conflicts showed two traits in QFP; When directly involved in a crisis, it tends to rely on Crises diplomacy, Multilateralism discourse, and intensifies hedging strategy regionally as well as globally. In crises where it is not directly implicated, mediation emerges as its principal foreign policy instrument. Political implications such as policy making, regional challenges, foreign policy response were discussed
Suitability of native tree species across regions of Rwanda in relation to climate sensitivity and ecosystem services
No full textOur ability to select native tree species for forest restoration and ecosystems services in tropical areas is limited by insufficient knowledge of tree growth, survival and climate sensitivity. This thesis aims to enhance the understanding of the suitability of tree species from highland tropical forest in current and future climates across Rwanda's agro-ecological regions through three experiments and one literature study. Two experiments with young trees were established at three sites along an elevation gradient (2400 to 1300 m a.s.l., temperature difference of 5.4 °C): one involving multispecies plantations of 20 species and another with potted trees of two species planted in the same soil at all sites. A third experiment, conducted in the Ruhande Arboretum (∼1700 m a.s.l.) studied shade tolerance of six species. All experiments included a mix of early (ES) and late (LS) successional species, with the multispecies plantations containing species with dominant distribution in different vegetation types and elevations (transitional rainforest at 1600 – 2000 and montane rainforest at > 2000 m a.s.l). The literature study assessed the suitability of 81 native tree species to different potential natural vegetation systems of east Africa and their contributions to different ecosystem services. The results from the multispecies plantation experiment showed that warming stimulates early tree growth in most early-successional species, particularly those originating from transitional rainforests. In contrast, several late-successional species, especially from higher elevations, did not respond or grew slower and had higher mortality at warmer sites. Findings from the potted tree experiment, although only limited to two species, aligned with results from the multispecies plantations, indicating that warming and not soil differences was the primary explanation of the observed site differences in tree growth in the larger study. In warmer climates, total biomass increased in the ES species without altering biomass allocation. In contrast, in the LS species, only root mass increased at warmer sites. The shading experiment revealed that dense canopy conditions significantly reduced the total tree biomass differently between species. However, under open sky, late-successional (shade-tolerant) species grew equally well as early-successional (shade-intolerant) species, suggesting tree growth differences in the elevation gradient were not due to species differences in light tolerance. The literature study of East African tree species showed clear relationships between elevation, climate, and species traits. Wood density increases as precipitation decreases, which it does with decreasing elevation. Among selected species, most provide multiple ecosystem services: 83% medicinal, 79% construction, 68% fuel, 58% edible, 56% cultural and 53% supporting and 31% regulating. The experimental studies suggest that in a warmer climate, higher-elevation and late-successional species may face increased competition from lower-elevation and early-successional species. When combined with findings from the literature, these results indicate that climate change will likely decrease the provisioning of ecosystem services specific to some late-successional species, as well as the biodiversity and carbon storage of Afromontane forests.
Key words: Native trees, Elevation gradient, Tree growth and mortality, Afromontane rainforest, Transitional rainforest, Successional group
MANDATE ADJUSTMENT AND THE PROTECTION OF CIVILIANS - An Analysis of UN Peacekeeping Operations in Eastern DRC
No full textCivilians are frequently victims of targeted violence during civil wars. As a response, the UNSC at times adapts the mandate of a UNPKO to explicitly include the protection of civilians. Despite this, it remains uncertain whether these targeted mandate adjustments result in better protection of civilians. To fill this gap, this thesis examines the influence of mandate adjustment to include the protection of civilians from physical violence on the protection of civilians in an ongoing civil war. The thesis argues that mandate adjustment influences the protection of civilians via organisational change. The inclusion of the protection of civilians as a task to be implemented triggers changes in the structure of the mission, which in turn necessitates internal and external coordination, hence improving the operational approaches of UN peacekeepers to demands for the protection of civilians from physical violence. To test this argument, I qualitatively analysed the case of UNPKO in the eastern DRC from 2000 to 2014 using theory-testing process tracing. While the findings suggest that mandate adjustment is likely to lead to a proactive operational approach by UN troops to the demand for the protection of civilians, the presence of a large, well-trained and well–equipped UN troop contingent does not necessarily deter belligerents from either continuing hostilities or targeting civilians. These findings, in addition to contributing to the existing literature on peacekeeping, challenge previous findings on the deterrence effect of the presence of a large, well-trained and well–equipped UN troop contingent on belligerents´ decisions to continue with hostilities and or target civilians
Non-Functional Requirements for Machine Learning Systems
No full textBackground: Machine learning (ML) systems are increasingly being deployed in complex and safety-critical domains such as autonomous driving, healthcare, and finance. ML systems learn using big data and solve a wide range of prediction and decision-making problems that would be difficult to solve with traditional systems. However, increasing use of ML in different systems has raised concerns about quality requirements, which are defined as non-functional requirements (NFRs). Many NFRs, such as fairness, transparency, explainability, and safety, are critical in ensuring the success and acceptance of ML systems. However, many NFRs for ML systems are not well understood (e.g., maintainability), some known NFRs may become more important (e.g., fairness), while some may become irrelevant in the ML context (e.g., modularity), some new NFRs may come into play (e.g., retrainability), and the scope of defining and measuring NFRs in ML systems is also a challenging task.
Objective: The research project focuses on addressing and managing issues related to NFRs for ML systems. The objective of the research is to identify current practices and challenges related to NFRs in an ML context, and to develop solutions to manage NFRs for ML systems.
Method: This research follows a design science methodology and consists of a series of empirical and design-oriented studies. First, we conducted an interview study to explore practitioners' perceptions of NFRs and the challenges associated with defining and measuring them in ML systems. Then we conducted a subsequent survey study to validate and expand these findings with broader practitioner input. To complement these studies, we conducted a partial systematic mapping study to assess the coverage of NFRs in the academic literature, revealing discrepancies between research focus and industrial needs. Additionally, we conducted group interviews with domain experts in the automotive industry to uncover requirements engineering (RE) practices and challenges specific to ML-enabled perception systems. Based on these insights, we proposed a structured, five-step quality framework and evaluated it through practitioner interviews. Finally, we proposed revised maintainability metrics adapted to the unique structure of ML systems, and we evaluated them using ten real-world open-source ML projects.
Findings: We found that NFRs are crucial and play an important role in the success of the ML systems. However, there is a research gap in this area, and managing NFRs for ML systems is challenging. To address the research objectives, we have identified important NFRs for ML systems, such as accuracy, reliability, fairness, transparency, retrainability, and explainability. We also identified challenges in defining, scoping, and measuring NFRs, including domain dependence, lack of standardized metrics, and difficulty in tracing NFRs across ML system components. Furthermore, we found that practitioners face significant challenges in applying RE to ML systems---particularly in autonomous perception—due to uncertainty, evolving components, and lack of systematic approaches for managing quality trade-offs, data quality, and cross-organizational collaboration. To address these gaps, we proposed a five-step NFR management framework, covering NFR selection, scoping, trade-off analysis, measurement planning, and structured specification using templates. Finally, given that maintainability is an important NFR for ML systems, we proposed scope-aware definitions and measurement strategies for maintainability in ML systems and demonstrated their usefulness through empirical evaluation.
Conclusion: NFRs are critical for ML systems, but they are difficult to define, allocate, specify, and measure due to challenges like unintended bias, non-deterministic behavior, and the high cost of thorough testing. Industry and research lack well-structured solutions to manage NFRs for ML systems effectively. This research addresses this critical gap by providing a comprehensive understanding of NFRs and the unique challenges they pose in the ML context. Through a combination of empirical studies and the development of a structured NFR management framework, this research offers a solution for identifying, prioritizing, scoping, measuring, and specifying NFRs across granular-level components of ML systems. Contributions also include scope-aware definitions and measurement metrics of maintainability for ML systems. These findings enrich the theoretical understanding of NFRs for ML systems, provide empirically grounded insights into their challenges, and introduce artifacts and metrics to support future research. These outcomes also provide valuable guidance for practitioners to build trustworthy, maintainable, and high-quality ML systems. This research will help practitioners make better engineering decisions, improve quality assurance processes, and provide a foundation for more systematic and accountable ML system development
Drivkrafter och strategier för fackligt engagemang - betydelsen av autonomi, kompetens och samhörighet för förtroendevalda och fackförbund
No full textThis study looks at what motivates people to take on union roles in Sweden and how unions
work to encourage such involvement. Our research is particularly relevant as Sweden has
seen a significant decline in the number of union representatives over time. Using
Self-Determination Theory, we explore how autonomy, competence, and relatedness affect
motivation. Through interviews with seven people, four union staff members and three
former representatives, we found that people join unions mainly to support shared values and
make positive changes, not for rewards. Having real influence greatly increases motivation,
while lack of time and lean workplaces are major obstacles. Unions focus on building skills
through training but don't provide enough support for members autonomy and relatedness,
even though this is vital for lasting engagement. Diminished power resources in the Swedish
labor market through declining union density, more insecure employment forms,
individualized wage systems, and weakened collective bargaining structures directly impact
members' motivation by limiting their experience of autonomy in their roles. This research
helps explain why fewer people are taking union roles in Sweden today and suggests how
unions might better address members' basic needs in today's changing work environment
Towards a System-Level Functional Language: Lithium
No full textFunctional programming has a rich and well-documented history. In functional programming, large problems can be described by the composition of smaller building blocks. Despite its benefits, functional programming has struggled to find its way into system-level programming. By leveraging the guarantees linear types impose, functional programming languages can be applied to system-level programming without sacrificing performance.
This thesis presents Lithium, a system-level functional programming language that is based on a variant of linear logic. Lithium is designed to be an intermediate compilation target for linear functional languages. We give the typing and kinding rules for Lithium before describing a series of transformations to turn Lithium into a language that is easily translated into assembly code. Additionally, we present a compilation scheme, a mapping from types to memory, and the application binary interface (ABI)
Investigating the Association Between Household Gender Dynamics, Intimate Partner Violence, Mothers' Mental Health, Child Abuse, and Child Stunting in Rwanda
No full textBackground: Many low- and middle-income countries still encounter the persistent challenge of child stunting, despite the widespread implementation of public health interventions aimed at reducing it. Most studies and interventions have primarily focused on child-related factors, often overlooking the broader household conditions. A deeper understanding of the household environment, including the health and behaviors of mothers and the family dynamics surrounding child-rearing, is crucial for addressing the stunting problem more effectively.
Aim: This thesis aims to investigate the association between household dynamics, such as gender roles, social support, women’s decision-making, intimate partner violence (IPV), mothers' behaviors related to child disciplining, and mothers’ mental health disorders, with stunting of children aged between one month and three years.
Methods: The studies in this thesis are based on a cross-sectional, population-based study design. Data was collected through interviews with mothers using an investigator-administered questionnaire, which included standardized WHO questions for assessing IPV, the Mini International Neuropsychiatric Interview questionnaire for evaluating mental health, and UNICEF Multiple Indicator Cluster Surveys (MICS)-guided questions for assessing child abuse. The Multidimensional Poverty Index (MPI), created by Oxford University, was used to evaluate the household’s socioeconomic status. In total, 601 mothers along with their children, who were between one month and three years old, participated in the study. Using a geographic grid method in collaboration with Lund University, villages in Rwanda’s Northern Province were identified, and households were selected through a simple random sampling process. Child age, along with weight and height measurements, served to compute height-for-age z-scores for assessing stunting, per WHO anthropometric guidelines. Descriptive measures and logistic regression models, both bi-variable and multivariable, were employed in the data analysis, showing adjusted odds ratio (aOR) and 95% confidence intervals.
Results: Among the 601 mother–child pairs, 27.1% of the children were stunted. Within the stunted group, 60.1% were boys and 39.9% were girls. Stunting increased with age, peaking at 39.9% in children aged 25–36 months. The mothers were predominantly from low-income backgrounds (88.4%), over 30 years old (56.4%), had low educational attainment (78.2%), and worked in unskilled jobs (97.3%). The average MPI was 0.265 across the sample. There was no statistical difference in the MPI scores between households with stunted children and those without stunted children. Nearly half (47.4%) of mothers endured IPV both prior to and during pregnancy. Mothers’ experience of physical IPV prior to pregnancy [aOR=1.29 (1.01 to 2.03)], along with sexual IPV encountered during pregnancy [aOR=1.25 (1.04 to 2.01)], were associated with child stunting. Depression was also a key risk factor, as children of mothers with current major depressive disorder had stunted growth [aOR = 1.67 (1.06–2.61)]. Both insufficient support while ill and lack of personal guidance of mothers were associated with child stunting, with [aOR=1.93 (1.13 to 3.28)] and [aOR=2.44 (1.41 to 4.26)], respectively. Harsh disciplining methods used by mothers, especially when both physical and non-physical approaches were employed, were linked to a higher odds ratio of child stunting [OR = 1.92 (1.08–3.41)].
Conclusions: Despite ongoing public health interventions in Rwanda, child stunting remains a significant problem. The findings highlight that child undernutrition is not only influenced by direct nutritional factors but is also deeply rooted in household dynamics, mothers’ well-being, and gendered social structures. Mental health disorders, exposure to IPV, lack of social support, and violent child-disciplining practices were all strongly associated with child stunting. Integrating mental health screening for pregnant women, educating expectant mothers on the importance of social support, and raising awareness about the dangers of violent child disciplining methods and encouraging non-violent alternatives are essential additions to the ANC content. These results underscore the need for a holistic, interdisciplinary, multisectoral approach that incorporates mental health services, IPV prevention, gender-sensitive interventions, enhanced ANC, and parental support programs into existing nutrition and child health initiatives. Tackling these broader determinants is essential for sustainable improvements in child growth and long-term public health gains
KOMMUNIKATION NÄR SINNENA SVIKTAR. En litteraturöversikt om kommunikationen mellan sjuksköterskor och personer med demenssjukdom
No full textBakgrund: Demenssjukdomar har inget botemedel och påverkar hela tillvaron för den
drabbade och personer i dennes närhet. Den kognitiva förmågan som innefattar att tolka och
förstå information, samt förmågan att göra sig förstådd är något sjukdomen präglas av.
Sveriges befolkning lever allt längre, vilket också innebär en ökning av förekomsten av
demenssjukdomar. Detta ställer krav på sjukvården och på sjuksköterskans bemötande samt
kommunikationsförmåga. Syfte: Att undersöka kommunikationen mellan sjuksköterska och
personer med demenssjukdom utifrån ett omvårdnadsperspektiv. Metod: Metoden är en
litteraturöversikt och innefattar litteratursökningar i databaserna PubMed och Cinahl. Artiklar
valdes utifrån relevans för syfte och genom krav på vetenskaplig kvalitet. 10 originalartiklar
valdes och dessa analyserades och samanställdes genom tillämpning av Fribergs
analysmodell (2022). Resultat: I resultatet framkom fyra huvudteman: Verbal
kommunikation, icke-verbal kommunikation, personen bakom sjukdomen och
närståendekommunikation. Resultatet visar vikten av att anpassa kommunikationen efter
individen samt att kombinera verbala- och icke-verbala verktyg. Att inkludera närstående
framkom också som viktigt för att underlätta kommunikationen. Slutsats: Effektiv
kommunikation med personer med demenssjukdom kräver en kombination av verbala- och
ickeverbala strategier samt ett personcentrerat förhållningssätt. Detta främjar förståelse,
trygghet och god omvårdnad och understryker behovet av relevant kunskap hos
sjuksköterskor
STATENS ANSVAR FÖR HISTORISKA UTSLÄPP AV VÄXTHUSGASER En normativ analys av analogin mellan slaveri och gamla utsläpp
No full textThis essay seeks to provide an answer to the subject of the state’s responsibility for historical emissions of greenhouse gases. This is done through a normative analysis of the analogy between slavery and “old” emissions. In this essay, I take the assumption that states sometimes are responsible for things that happen on their land. Afterwards I evaluate an argument based on two premises. The first says that the state has responsibility for historical injustices like slavery. The second premise says that the state’s responsibility for slavery is equally relevant to the state’s responsibility for old emissions. I argue that both premises are true, and that the state therefore is responsible for old emissions. This is based on the intuition that its fair that the state is responsible for things like slavery and need to compensate for this. I further argue that reparations for slavery and old emissions implicate ethical collectivism, and that the state is the only agent that can provide victims with the reparations needed. At last, the result shows how a backward-looking principle like the polluter pays principle can be interpreted in a collective way
Philadelphia-Negative Myeloproliferative Neoplasms: Diagnostic Insights, Genetic Factors and Vascular Impact on Survival
No full textThe thesis addresses Philadelphia-negative myeloproliferative neoplasms (MPNs), a group
of clonal hematopoietic stem cell disorders encompassing polycythemia vera (PV), essential
thrombocythemia (ET), myelofibrosis (MF) and MPN Unclassifiable (MPN-U). These diseases,
characterized by somatic mutations in JAK2, CALR, and MPL, involve diagnostic
challenges, variable risk of vascular complications, and diverse survival outcome. The studies
presented in this dissertation aim to enhance the understanding of genetic predisposition
to treatment outcome, survival, and vascular complications associated with MPNs.
The first study examines genetic variations in IL28B (IFNL3) and their influence on interferon-
alpha therapy outcomes, demonstrating the potential for genetic markers to predict
treatment efficacy. The second and third studies, based on data from the Swedish MPN
registry, investigate survival patterns and vascular complications in patients with PV, ET
and MF. These finding emphasize the prognostic significance of vascular events and the
protective role of cytoreductive therapies. The fourth study evaluates the heterogeneity of
patients with MPN-U, addressing diagnostic challenges and their implications for classification
and clinical management while documenting survival patterns and the incidence of
thrombotic complications.
Together, these studies enhance the understanding of MPNs by addressing diagnostic challenges
and examining clinical and therapeutic outcomes. The findings underscore the need
for tailored approaches in diagnosis, risk stratification, and treatment to improve patient
care and provide a foundation for further research into these complex disorders