Blekinge Institute of Technology
Not a member yet
13576 research outputs found
Sort by
Evaluating the Quality of GenAI Applications in Software Engineering : A Multi-case Study
No full textContext: Generative AI (GenAI) is increasingly adopted in software development for tasks such as document generation, data analysis, and code generation.However, evaluating the quality of GenAI applications becomes challenging, as traditional quality measurements may not be fully applicable. Objective: In this study, we explore how practitioners evaluate the quality of GenAI applications and investigate quality evaluation techniques. Method: We conducted a multi-case study in three industrial projects from software development companies.We examined four GenAI application domains: document generation, data analysis and insight generation, customer service, and code generation.Data were collected through three workshops and 23 semi-structured interviews with industrial practitioners. Results: We identified fourteen GenAI use cases and 28 metrics currently used to evaluate the quality of GenAI applications' outputs.We synthesized the identified metrics' usage patterns and challenges based on the collected data. Conclusions: This study presents practical insights into using metrics to measure GenAI-based system qualities in real industrial settings.Our findings indicate that practitioners use custom-built and context‑specific metrics; combining these with academic metrics can strengthen GenAI system quality evaluation
Quality Evaluation of Generative AI Systems : Processes, Metrics, Methods, and Frameworks for Industrial Software Engineering
No full textGenerative Artificial Intelligence (GenAI) is being rapidly adopted in software engineering, introducing a paradigm shift toward human-AI co-creation. However, the non-deterministic, probabilistic, and often black-box nature of GenAI models presents challenges for traditional software quality assurance. Conventional verification and validation techniques are insufficient to handle outputs that are neither predictably correct nor incorrect, but rather stochastically plausible. This discrepancy creates an urgent need for practical processes, metrics, and new governance frameworks to evaluate and manage the quality of GenAI systems in industrial environments.This thesis examines how industrial organizations adopt GenAI, identify metrics, and evaluate system qualities in alignment with ISO quality standards. Case studies were employed to explore real-world adoption processes, identify context-specific industrial metrics, and uncover practical insights within organizations. A snowballing literature review was conducted to systematically identify, categorize, and synthesize academic metrics for evaluating the output of GenAI systems. Finally, a controlled experiment was designed to quantitatively test the efficiency (e.g., E2E generation time) and effectiveness (e.g., accuracy) of GenAI agent choices. The main contributions of this thesis are a synthesized actionable model and framework grounded in both industrial practice and quality standards. The first contribution is a four-stage adoption model, denoted as the IMRM model (Innovate → considerations, Measure → metrics, Realize → values, Manage → improvements) that integrates early-stage risk assessment (e.g., legal, security, and licensing) andquality evaluation throughout the GenAI adoption and usage.The second contribution presents a detailed framework that connects risks andmetrics to concrete decision support, justifying the business value (e.g., quality gates) and technical trade-offs of GenAI solutions. The third contribution provides a structured mapping of GenAI quality to ISO/IEC 25010, 25023, and 25059 characteristics, attempting to ground practical evaluation needs within a standardized vocabulary. This thesis concludes that a structured quality evaluation process, which prioritizes risks and context, is a valuable approach intended to support building the business confidence required to leverage GenAI for efficient and effective software engineering in industry
SoK : Evolution of the Key Encapsulation Mechanism's Role in Cryptographic Migrations for IoT Systems
No full textKey Encapsulation Mechanisms (KEM) is a special case of Public Key Encryption (PKE) that was recently standardized by National Institute of Standards and Technology in USA. The broader adoption of the term in industry practice was necessitated by the discovery of the malleability property of ciphertext, which led to new approaches PKE. New standard initiated refactoring of all cryptographic software libraries and this process relates to the problems of cryptographic agility. This Systematization of Knowledge (SoK) addresses the developments of public key encryption methods and the main challenges that drive the specialization of KEM in cryptographic software. Based upon our findings from a systematic literature review, we present a formal analysis that provides cryptographic users a means of better understanding of KEM and its roles in cryptographic migrations in IoT systems. We have identified the main milestones of KEM evolution and structured it into four development areas. We found that the evolution of KEM is defined by a variety of mathematical foundations that always reflect various aspects of the cryptosystem. Our findings indicate that academia, industry practitioners and standardization bodies propagate such approaches into practice by additional abstraction layers in cryptographic software libraries. However, the libraries is still not in consensus, which is confirmed after the discovery of a new class of libraries, cryptographic bindings. To structure the mentioned phenomena, we introduced a novel, three-facet, consumer-centered mapping of the data security domain.We believe our contribution can help researchers and practitioners to have a broader and deeper understanding of data encryption tooling in context of cryptographic migrations.SERIC
Life Events Preceding Alopecia Areata Onset—a Descriptive Retrospective Cohort Study Focusing on Strain
No full textBackground: Patients often relate preceding life events to the onset of alopecia areata (AA). This cohort study systematically examined number and type of life events together with experienced strain during two years before AA onset. Methods: The study involved 100 patients with AA who reported life events through the Social Readjustment Rating Scale (SRRS). Life event strain was examined using the Streiner model, a stress-based model in which strain is defined as undesirable, unanticipated and uncontrollable. SRRS readjustment rating and the rated strain for each life event were obtained. Percentages of strainful life events for each SRRS category were calculated. Associations between AA disease-variables, sociodemographic-variables, Beck Depression Inventory-scores and Beck Anxiety Inventory-scores were examined in relation to the number of SRRS events and the number of strainful events. Results: Total life events reported were median (range) 7.0 (0–22), almost two times higher than lifetime population studies on life events. The most experienced life events were loss of health and separation. At least one strainful life event prior to the AA onset was reported by 82% of patients. The highest strainful ratings were found in the Family and Personal categories of SRRS, such as Major change in health of a family member and Major personal injury or illness. A comparison between SRRS life events and strain ratings showed that 60% of the life events were rated higher and 35% were rated lower than the standard SRRS readjustment weight order. Conclusion: This study provides support for an inquiry into any strainful life event preceding AA onset in newly diagnosed patients with AA. Categories of highly strainful events and the importance of subjective strain are identified in patients with AA. This implies the need to address strain, which seems to be important for strain reduction thereby decreasing the burden among patients with AA.
How to enhance continuity of strategic sustainable development work in municipalities and regions - findings from Sweden and Finland
No full textDespite numerous initiatives, methodologies and approaches, municipalities and regions continue to face challenges in sustaining long-term efforts toward sustainability. While continuously learning from mistakes and success factors, comprehensive recommendations for how to take a long-term approach for the work, specifically developed for leadership in local and regional contexts, are largely lacking. Therefore, the purpose of this study was to identify a comprehensive set of essential factors that could support long-term sustainability efforts. The study is mainly related to the Framework for Strategic Sustainable Development, but the findings are likely relevant also in relation to other comprehensive sustainability methodologies and approaches. Re-analysis of longitudinal data from action research in Sweden and Finland is supplemented by new qualitative data collection, the study identifies eight conditions for continuity. Being introduced to these conditions upfront may save valuable time and could be utilized as a checklist to be revisited regularly and to strive toward
Privacy by design : Aligning GDPR and software engineering specifications with a requirements engineering approach
No full textContext: Consistent requirements and system specifications are essential for the compliance of software systems towards the General Data Protection Regulation (GDPR). Both artefacts need to be "grounded" in the original text and conjointly assure the achievement of privacy by design (PbD). Objectives: There is little understanding of the perspectives of practitioners on specification objectives and goals to address PbD. Existing approaches to GDPR and PbD do not account for the complex intersection between problem and solution space expressed in GDPR. In this study we explore the demand for conjoint requirements and system specification for PbD and suggest an initial version of an approach to address this demand. Methods: We reviewed existing secondary and related primary studies on GDPR compliance and conducted interviews with practitioners to (1) investigate the state-of-practice in requirements and system specifications for GDPR compliance and (2) understand the underlying specification objectives and goals (e.g., traceability). We developed and evaluated an initial version of an approach for requirements and systems specification for PbD, and evaluated it against the specification objectives. Results: The relationship between problem and solution space, as expressed in GDPR, is instrumental in supporting PbD. We demonstrate how our approach, based on the modeling GDPR content with original legal concepts, contributes to specification objectives of capturing legal knowledge, supporting specification transparency for roles involved, and traceability. Conclusion: In addition to assuring traceability, GDPR demands need to be addressed throughout different levels of abstraction in the engineering lifecycle to achieve PbD. Legal knowledge specified in the GDPR text should be captured in specifications to address the demands of different stakeholders and ensure compliance. While our results confirm the suitability of our approach to address practical needs, we also revealed specific needs for the future effective operationalization of our suggested approach
Mapping Fake News Research in Digital Media : A Bibliometric and Topic Modeling Analysis of Global Trends
No full textThis study aims to identify research trends in communication regarding the phenomenon of “fake news” in digital media. Fake news has become a rapidly growing and significant area of research in communication studies in recent years. Published studies were collected from the Science Citation Index Expanded database. The analysis included the annual distribution of publications, citation metrics, leading journals, countries, institutions, and authors. To explore the conceptual structure, topic modeling was conducted using text mining techniques along with DBSCAN and k-means clustering methods. The United States is a leader in the field, both as a producer country and in terms of technology implementation. Vraga, Bode, Tully, Hameleers, and Tandoc are among the most influential authors. The most cited studies specifically focus on misinformation in the health sector and political disinformation and manipulation during elections. Topic modeling analyses show that the literature mainly clusters around health disinformation, political communication, and verification technologies. The findings have important implications for communication policies, media literacy, and fact-checking technologies. Research that systematically examines fake news from a communication perspective, both in performance and conceptual structures, is scarce in the literature. The resulting thematic clusters provide valuable insights for future research.
The Effect of Evening Technology Use on Objective Sleep in Older Adults : Protocol for a Crossover Randomized Controlled Trial
No full textBackground: Evening technology use (ETU) has been associated with sleep disturbances, often attributed to blue light exposure and cognitive arousal. However, most of the existing evidence focuses on younger populations and relies primarily on subjective measures. As older adults increasingly engage with both passive and active technology use, it is important to investigate how ETU impacts objective sleep. Currently, there is also a limited understanding of how particular evening digital activities, especially active versus passive engagement, affect objective sleep in older adults. Objective: This study aims to investigate the impact of exposure to ETU on both objective and subjective sleep outcomes in older adults. Methods: This is a randomized crossover trial involving approximately 55 adults aged 60-75 years from the ongoing Swedish National Study on Aging and Care-Blekinge. Each participant will undergo 3 one-week intervention periods: active ETU, passive ETU, and a nondigital activity (book reading), with one-week washout periods in between. The order of interventions will be randomized. Sleep will be assessed using a home-based electroencephalography device (MUSE headband) and daily self-reports. Primary outcomes are sleep onset latency and wake after sleep onset. Secondary outcomes include objective measures such as total sleep time, sleep efficiency, and time spent in REM, deep, and light sleep, subjective sleep quality, adherence, and perception of the intervention and comfort of using the objective measurement tool, that is, the electroencephalography headband. Linear mixed-effects models (with fixed effects for condition and period and a random participant intercept) were used to analyze crossover effects on sleep outcomes. Results: Participant recruitment and data collection began in the fall of 2025 and will continue through summer 2026 or until the target sample size is reached. Data collection is scheduled to be completed by spring 2027. Results will include participant flow, baseline characteristics, adherence data, and comparative analyses of the 3 intervention conditions. Within-subject statistical models will be used to evaluate differences in sleep outcomes and investigate the associations between ETU and sleep quality. Conclusions: This crossover study will clarify how active and passive ETU, compared with a nondigital activity, relate to objective sleep in older adults. Findings will inform simple, practical recommendations for technology use before bed in late life.
Classification of saturated Fell bundles : The discrete case and beyond
No full textWe present a classification framework for saturated Fell bundles over groups, utilizing data associated with their base group and unit fiber. This framework provides a unified perspective on the structure and properties of such bundles and yields key insights into their classification. In the case of discrete groups, we obtain a complete and transparent classification in terms of generalized factor systems, leading to a cohomological description of equivalence classes. For general locally compact groups, the situation is more delicate: While our construction extends to this setting, the classification depends on choices of topology on the underlying Banach bundle
Value Stream Mapping : Introducing a Lean Production Method in Consulting Firms
No full textValue Stream Mapping (VSM) is a well-established Lean method for mapping flow of value and identifying waste in manufacturing contexts. The application of VSM within service and knowledge-intensive environment remains underexplored. Consulting processes are characterized by low standardization (due to case dependency), variability and heavy reliance on customer interaction. This poses a challenge for traditional VSM application. The purpose of this study is to explore how to apply VSM in the consulting context to identify the flow in value, improvement opportunities, inefficiencies of workflow and how a Future-state Value Stream Map (FVSM) can be established and juxtaposed with a current-state VSM. The study applies an exploratory research design that combines qualitative and quantitative collection and analysis methods through a case study. The quantitative data was collected through the business case using descriptive process analysis to examine different timestamps, distributions of outcomes and rates of success. The qualitative data was collected through semi-structured interviews with consultants and their managers. The qualitative data was analyzed using a deductive qualitative content analysis (DQICA) to adapt to analyzing the consulting work process through VSM. The results show that mapping of value creation (VC) in consulting is primarily in the delivery phase. The non-value adding (NVA) waste was identified to reside during the pre-contract phases, due to waiting times, rework and repeated customer interaction. The FVSM showcases how potential improvements can lower wastes by nearly 9% through estimations by consideration of previous findings. The study concludes that VSM possibly can be meaningfully applied in the consulting field through adaptations to service oriented conditions.Value Stream Mapping (VSM) är en väletablerad Lean metod som används för att identifiera slöseri och förbättra värdeflöden inom tillverkningsindustrin. Tillämpningen av VSM är inte etablerad inom kunskapsintensiva och tjänsteriktade miljöer som exempelvis konsultverksamheter. Konsultprocesserkännetecknas av hög variation, låg grad av standardisering och ett starkt beroende av kundbeteende. Dessa aspekter utmanar hur VSM har använts historiskt sätt. Syftet med detta examensarbete är att utforska hur VSM kan tillämpas i ett konsultföretag för att identifiera förbättringsmöjligheter och ineffektiva processer. Vidare visas det hur en framtida värdeflödeskarta (Future State Value Stream Map, FVSM) kan sammanställas och jämföras med en VSM. Examensarbetet har en explorativ forskningsdesign och kombinerar kvalitativa- samt kvantitativa datainsamlingsmetoder. Den kvalitativa data samlades in genom semistrukturerade intervjuer med konsulter och chefer. Deras perspektiv tolkades och analyserades med hjälp av en deduktiv kvalitativ innehållsanalys (Deductive Qualitative Content Analysis, DQICA) för att anpassa VSM metodiken till konsultverksamhetens arbetsflöde. Den kvantitativa data som insamlades från systeminformation från affärsärenden, vilket analyserades med en deskriptiv processanalys för att undersöka statusfördelningar och väntetider mellan faser. Resultatet visar att värdeskapande processen är huvudsakligen grundad i leveransfasen, där icke-värdeskapandet befinner sig främst i tiden före kontrakt skrivs på mellan kund och konsult. De icke-värdeskapande är i identifierat att orsakas av väntetider, omarbeta och repetitiva interaktioner mer kunder.FVSM illustrerar genom teoretisk tolkning att potentiella förbättringar hade kunnat leda till 9% minskningar i icke-värdeskapande vilket etablerades från tolkning av tidigare studier. Studien dras i slutsatsen att VSM kan tillämpas meningsfullt inom konsultbranschen endast när metoden anpassas till tjänstesärskilda förutsättningar. Vidare kan metodiken av denna studie användas som ett analytiskt tolkningsramverk för framtida utforskning