Blekinge Institute of Technology
Not a member yet
13576 research outputs found
Sort by
Predicting Depression in Older Adults : A Novel Feature Selection and Neural Network Framework
No full textDepression in older adults is a significant public health issue with broad impacts on both individuals and society. The multifaceted nature of depression underscores the complexity of identifying and predicting risk factors, necessitating a sophisticated and accurate approach based on new emerging technologies. Compared to traditional statistical methods, machine learning provides a more detailed and individualized understanding of risk variables by analyzing large datasets, identifying patterns, and building predictive models. This study presented a novel feature selection method based on the relief and lasso algorithms. The proposed feature selection method selected the ten most significant features from the dataset. A neural network (NN) with hyperparameters optimized by a grid search technique was used to categorize depression. The feature selection and classification modules work together as a single unit, namely as (Relief_Lasso_NN). Data from the Swedish National Study on Aging and Care (SNAC) was used for this study. The collected dataset consists of 726 samples with 75 features per sample. Four experiments were conducted to validate the performance of the proposed (Relief_Lasso_NN) framework. The proposed model achieved an accuracy of 90.34% in predicting depression using only ten features from the dataset. The top 10 features identified by the proposed feature selection method significantly impact depression in older adults. Furthermore, the performance of seven other state-of-the-art machine learning models was also compared with the proposed framework.SNA
Understanding the role of Key Encapsulation Mechanisms in Cryptographic Migrations : Towards Cryptographic-Agility in IoT Systems Based on End-to-End Encryption Approach
No full textThe increasing data-security regulation and cyber-threats requires IT vendors to use new cryptographic tools or refactor live systems to encrypt existing data. One step in that direction is integrating an appropriate security protocol and cryptographic software library during the system design phase. However, it is not sufficient when it happens to migrate existing data to encrypted form in the live system on-the-fly, re-encrypt data to a different encryption standard, or have the data of the same origin but encrypted with different standards. This thesis explores the new emergent area of cryptographic agility, which focuses on various challenges while adopting cryptographic migrations in live systems. We proposed End-to-End Encryption (E2EE) design for telemetry data security in two different applications: maritime surveillance and drone-management. We aimed to understand the role of Key Encapsulation Mechanism (KEM) cryptographic primitive in the data-security domain. The notion of crypto-agility constitutes a context-sensitive, activity-based perspective on data security. In this thesis, we aim at both understanding and exploring practical possibilities of this notion. We employ a mixed-methods approach to achieve our aim: Experimentation, Literature Review and Survey. We have studied and applied quantum-safe KEM cryptographic primitives to simulate practical cryptographic migration in live IoT systems. We have shown the importance of KEM security properties and the performance of KEM primitives for telemetry data confidentiality. We proposed new crypto-agility values and trade offs as decision making support tool for consumers of cryptographic technologies. Furthermore, we have employed systematization of knowledge to structure how different types of contributions developed various KEM notions, its influence on the standardization process, and presence in cryptographic software libraries over the last 40 years. The proposed approaches have been shown to be capable of explaining the role of KEM in cryptographic migrations and underlying properties of crypto agility. This can facilitate domain experts in narrowing down the scope of analysis while achieving sufficiency for cryptographic migrations in live IoT systems based on end-to-end encryption protocols.Connect2SmallPort
Etik i algoritmernas era: Sjuksköterskans roll i en AI-driven vård : En litteraturstudie som beskriver hur etiska aspekter påverkar sjuksköterskans arbete vid användning av AI inom vården
No full textBakgrund: Användningen av Artificiell intelligens (AI) i hälso- och sjukvården har ökat kraftigt. Sjuksköterskan spelar en central roll i implementeringen, särskilt inom den patientnära vården. AI väcker etiska frågor om autonomi, ansvar och patientsekretess, där kunskapsluckor fortfarande finns. Syfte: Syftet med studien var att beskriva hur de etiska aspekterna i sjuksköterskans arbete påverkas vid användning av AI inom vården. Metod: En allmän litteraturöversikt med induktiv ansats genomfördes som metod. Vetenskapliga och peer-reviewed artiklar på engelska publicerade mellan 2019 och 2025 inkluderades. Både kvalitativa och kvantitativa studier valdes utifrån tydliga inklusions- och exklusionskriterier samt avgränsningar. Resultat: Totalt granskades sju studier, vilka analyserades samt kategoriserades. Fem kategorier identifierades: bristande autonomi och minskat ansvar, etiska utmaningar kring patientsekretess och dataskydd, förändrad vårdkvalitet och mänsklig närvaro, etiska dilemman i det kliniska beslutsfattandet samt brist på reglering och policy skapar etisk osäkerhet. Sjuksköterskor upplevde att AI kunde förbättra vårdkvaliteten och lätta administrativa bördor, men riskerade samtidigt att minska sjuksköterskans autonomi och ansvar. Osäkerhet kring dataskydd och sekretess identifierades som etiska utmaningar. Brist på kunskap om teknologin och tydliga riktlinjer kunde försvåra etisk reflektion och ansvarstagande. Fynden pekade på behov av utbildning, sjuksköterskors delaktighet i utvecklingen av AI samt stödjande riktlinjer för att kunna integrera teknologin på ett etiskt hållbart sätt utan att ersätta sjuksköterskan. Slutsats: Användningen av AI inom vården påverkar sjuksköterskans etiska ansvar och yrkesroll. Teknologin kan både stödja och utmana det kliniska omdömet. För att AI ska stärka snarare än underminera omvårdnadens etiska värden krävs utbildning, tydliga riktlinjer och att sjuksköterskor ges möjlighet att utöva sitt professionella ansvar i relation till teknologin. Etisk integrering av AI är avgörande för att sjuksköterskor med sitt etiska ansvar ska kunna erbjuda personcentrerad vård med hjälp av väl anpassade digitala verktyg
Exploring the role of project status information in effective code smell detection
No full textRepairing code smells detected in the code or design of the system is one of the activities contributing to increasing the software quality. In this study, we investigate the impact of non-numerical information of software, such as project status information combined with machine learning techniques, on improving code smell detection. For this purpose, we constructed a dataset consisting of 22 systems with various project statuses, 12,040 classes, and 18 features that included 1935 large classes. A set of experiments was conducted with ten different machine learning techniques by dividing the dataset into training, validation, and testing sets to detect the large class code smell. Feature selection and data balancing techniques have been applied. The classifier’s performance was evaluated using six indicators: precision, recall, F-measure, MCC, ROC area, and Kappa tests. The preliminary experimental results reveal that feature selection and data balancing have poor influence on the accuracy of machine learning classifiers. Moreover, they vary their behavior when utilized in sets with different values for the selected project status information of their classes. The average value of classifiers performance when fed with status information is better than without. The Random Forest achieved the best behavior according to all performance indicators (100%) with status information, while AdaBoostM1 and SMO achieved the worst in most of them (> 86%). According to the findings of this study, providing machine learning techniques with project status information about the classes to be analyzed can improve the results of large class detection. © The Author(s) 2024
Understanding Remote Work Experience : Insights Into Well-Being
No full textBackground: After the pandemic, software engineers were forced to work remotely, in many cases without prior experience of doing so. Objective: The objective of this work is to analyze the factors that influence engineers' motivation, stress and performance when working remotely after the pandemic, and to what level. Methods: A significant number (around 1000) of Latin-American software development professionals from different countries who work remotely were surveyed in order to study the factors that affect them and how when they work in this manner. The data collected from the survey were then statistically analyzed using the partial least square-structural equation modeling (PLS-SEM) method. Conclusions: The analysis of the data made it possible to conclude that there are direct negative effects of stress on performance and direct positive effects of motivation on performance. In addition, we found that skills, experience, and teamwork behavior, such as trust, communication, and knowledge sharing, play an important role when working remotely.
Rapid improvement in a resource-constrained environment : Assessment with the iFLAP framework and improvement planning
No full textBackground. Resource-constrained environments are commonly found in software engineering and mean limited staffing, budgets, and more limitations, making it challenging to stay competitive. Software process improvement, SPI, is a way to improve the situation, but traditional SPI has been shown not to fit smaller enterprises and other environments where resources are constrained. Inductive assessment frameworks, such as iFLAP might be more suitable since their feature is to find improvements based on the individual situation. Objectives. This study aims to find if the iFLAP framework could be used in microenterprises, a special case of a resource-constrained environment with few staff and teams. One goal is to find how much effort the assessment with iFLAP requires. This research should also compare improvements in the software development process proposed by one person with improvements found by the iFLAP process. The final goal is to evaluate the effects of implementing an improvement plan based on the iFLAP assessment. Methods. An action research was performed in two iterations where the first iteration was to assess the current situation and interview the participants to elicit improvement categories. Then the improvements were validated through data triangulation, prioritized, and identified dependencies. The next iteration of the research was to plan improvement, and then implement a part of the plan for one month of evaluation. Finally, the participants had a short retrospective interview. Results. The assessment took 38 hours and identified the most important improvement areas: testing method, task management, and process improvement. Compared to that, a one-person proposal yielded the improvement areas of problem tracking, test management, and prioritization. Finally, the effects noted from the evaluation of the improvement plan were that communication had increased and there was a perception that the improvement plan was relevant to the process. However, involving whole the team and having short concise steps are crucial improvements to the proposed improvement strategy. Conclusions. iFLAP might be suitable for performing improvement assessments in resource-constrained environments. It should be noted that it requires a continuous effort for a longer time to notice significant changes. The importance lies in that the process should be improved in small steps, and people, motivation, and collaboration are vital for its success
Psykiska påfrestningar hos sjuksköterskor under COVID-19-pandemin
No full textBakgrund: Covid-19 pandemin startade december 2019 och pågick fram till början av juni 2023. Viruset spred sig snabbt över världen och blev snabbt klassad som en pandemi. Sjukvården runt om i världen blev fort påverkad och bristen på information skapade en stor oro bland befolkningen. Virusets varierande symptom skapade chock och förvirring även bland sjukvårdspersonalen. Jobbet som sjuksköterska medförde plötsligt försvårade arbetsförhållanden, karakteriserade av längre arbetstider och högre arbetsbelastning. Pandemin förde även med sig en ökad psykisk påfrestning för sjuksköterskor, vilket särskilt märktes genom en ökning av både ångest och depression. Syfte: Syftet med studien var att beskriva sjuksköterskors upplevelser av psykiska påfrestningar under covid-19 pandemin. Metod: En metod med kvalitativ design och en integrerad sammanställning av artiklar inspirerad av metasyntes. Resultat: Resultatet är uppbyggt av upplevelserna av psykisk påfrestning hos sjuksköterskorna. I resultatet framkom tre huvudkategorier av upplevelser: oro, ensamhet och arbetsmiljöns påverkan. Oro och arbetsmiljöns påverkan delades sedan in i två underkategorier av rädsla respektive utmattning. Slutsats: Sjuksköterskor upplevde att psykisk påfrestning under covid-19 pandemin var ett faktum. Ökad oro, rädsla och utmattning bland sjuksköterskor var ett resultat av den psykiska påfrestningen som de utsattes för. Bristen på information och kunskap var ett ständigt återkommande problem för personalen. Smittspridning och omhändertagandet av smittsamma patienter var en bidragande faktor till att sjuksköterskorna mådde psykiskt dåligt under pandemin.
Kontinuerlig Motivation i Förändringsarbete : En studie av förändringsarbete inom rättspsykiatrisk vård
No full textBakgrund: Förändringsarbete inom rättspsykiatrisk vård innebär särskilda utmaningar, därmedarbetares motivation ofta är avgörande men otillräckligt belyst. Trots stora satsningar på utvecklinginom vård, saknas det kunskap om vilka motivationskällor som bibehåller motivation hos medarbetarei förändringsarbeten inom komplexa vårdmiljöer. Syfte: Syftet med studien är att med Self-Determination Theory (SDT) undersöka bakomliggandeförutsättningar för att medarbetares motivation i förändringsarbeten ska bibehållas inom rättspsykiatriskvård. Metod: Studien har en kvalitativ ansats och bygger på semistrukturerade intervjuer med personerverksamma inom rättspsykiatrin som deltagit i förändringsarbete. Intervjumaterialet analyseradestematiskt med stöd av SDT för att identifiera återkommande mönster kopplade till motivation. Resultat. Flera motivationskällor identifierades, bland annat personligt driv, upplevda resultat,möjlighet till påverkan samt återkoppling och tydliga målbilder. Dessa aspekter bidrog till att stärka ochvidmakthålla motivation över tid. Slutsatser. Studien visar att flera inre motivationskällor beror på känslan av kompetens, autonomioch samhörighet, vilket kompletteras av yttre motivationskällor såsom ledningens instruktioner ochförväntningar. Förståelse om dessa parametrar är vitala för ledare att förstå under förändringsarbetensskridning. Background. Change initiatives within forensic psychiatric care present specific challenges, withstaff motivation often being a crucial yet underexplored factor. Despite significant investment in thedevelopment of healthcare services, there is limited knowledge regarding the factors that help sustainstaff motivation during organisational change in complex care environments. Objectives. The purpose of the study is to use Self-Determination Theory (SDT) to examine theunderlying conditions for maintaining employees’ motivation in change efforts within forensicpsychiatric care. Methods. This qualitative study is based on semi-structured interviews with professionals workingin forensic psychiatry who have participated in change initiatives. The interview material wasthematically analysed with the support of SDT in order to identify recurring patterns related tomotivation. Results. Several motivational factors were identified, including individual drive, perceivedprogress, opportunity for influence, as well as feedback and clearly defined goals. These factorscontributed to strengthening and maintaining motivation over time. Conclusions. The study demonstrates that intrinsic motivational factors are closely associated withthe experience of competence, autonomy, and relatedness, and are complemented by extrinsic factorssuch as leadership instructions and expectations. An understanding of these factors is essential forleaders managing and supporting organisational change processes.
From bearings to substations : Transfer Learning for fault detection in district heating
No full textFault Detection and Diagnosis (FDD) in District Heating (DH) systems is vital for improving operational efficiency. As DH networks evolve towards Fourth-Generation District Heating (4GDH), their reliance on lower operational temperatures intensifies the need for robust FDD. However, implementing effective FDD faces challenges due to the lack of labelled fault data and the complexity of DH substations. This paper introduces a novel FDD methodology using Transfer Learning (TL) to bridge the gap between faulty bearings, controlled DH substation experiments and real-world operational data. We propose a fault signature method that aligns the data to reduce the domain gap, revealing similarities between faulty bearings’ vibration patterns and ΔT readings in DH substations. The TL-enhanced models demonstrated robust performance, achieving F1 scores up to 98% on lab data and 91% on real-world operational data, respectively. These results mark a notable advancement in FDD of DH substations, as our method offers accurate fault detection and valuable insights across diverse operational contexts, ranging from valve issues, faulty sensors, wrong control strategies and normal behaviour. Notably, temperature dynamics resemble behaviour akin to faulty bearing vibrations, highlighting their potential as a critical indicator of a faulty substation, enabling more effective FDD in DH systems
Towards Structured Data Quality Assessment for Smart Grid SCADA-AI Pipelines : A Preliminary Exploration using a Graph-Based Approach
No full textEnsuring the quality of input data is essential for building robust and explainable AI models in critical infrastructure domains such as smart grids. However, in SCADA-based intrusion detection pipelines, structural inconsistencies and latent feature drift are rarely assessed. In this preliminary study, we adapt the DQuaG framework—a graph-based reconstruction model originally developed for general tabular data—to assess data quality in a SCADA dataset based on the DNP3 protocol. Weapply the model in an unsupervised setting, using reconstructionloss to detect potential inconsistencies without labeled errors. Our results show that even within clean datasets, structural outliers can be identified, highlighting the value of structure-aware data validation. We discuss the implications for data-centric AI pipelines in SCADA cybersecurity and outline future directions for improving quality assessment and synthetic data generation.This paper was presented as a poster at the 20th Swedish National Computer Networking and Cloud Computing Workshop (SNCNW 2025). The workshop does not publish formal proceedings, and the authors retain copyright of their contributions.</p