Blekinge Institute of Technology
Not a member yet
13576 research outputs found
Sort by
Identification of Spambots and Fake Followers on Social Network via Interpretable AI-based Machine Learning
No full textSocial networking platforms like X (Twitter) serve as hubs for open human interaction, but they are also increasingly infiltrated by automated accounts masquerading as human users. These bots often engage in activities such as spreading fake news and manipulating public opinion during politically sensitive times like elections. Most of the current bot detection methods rely on black-box algorithms, raising concerns about their transparency and practical usability. This study aims to address these limitations by developing a novel methodology for the detection of spambots and fake followers using annotated data. To this end, we propose an interpretable machine learning (ML) framework, leveraging multiple ML algorithms with hyperparameters optimized through cross-validation, to enhance the detection process. Furthermore, we analyze several features and provide a unique feature set that is optimized to offer excellent performance for bot detection. Moreover, we utilize multiple interpretable AI techniques which include Shapley Additive Explanations (SHAP) and Local Interpretable Model-agnostic Explanations (LIME). SHAP will help to display the effects of particular characteristics on the model’s prediction which will help in determining whether an account is a bot or a legitimate user. LIME will help to comprehend the model’s predictions, offering clarity regarding the traits or attributes that drive the classification conclusion. LIME allows researchers to detect bot-like activity in social networks by generating locally faithful explanations for each prediction. Our model offers enhanced interpretability by clearly highlighting the impact of various features used for spam and fake follower detection when compared to existing state-of-the-art social network bot detection systems. The results showcase the model’s ability to identify key distinguishing attributes between bots and legitimate users which offers a transparent and effective solution for social network bot detection. Additionally, we utilize two comprehensive datasets including Cresci-15 and Cresci-17, which serve as robust baselines for comparison. Our model showcases its effectiveness by outperforming other methods while providing interpretability which increases performance and reliability for the task of bot detection.
A proposal and assessment of an improved heuristic for the Eager Test smell detection
No full textContext: The evidence for the prevalence of test smells at the unit testing level has relied on the accuracy of detection tools, which have seen intense research in the last two decades. The Eager Test smell, one of the most prevalent, is often identified using simplified detection rules that practitioners find inadequate. Objective: We aim to improve the rules for detecting the Eager Test smell. Method: We reviewed the literature on test smells to analyze the definitions and detection rules of the Eager Test smell. We proposed a novel, unambiguous definition of the test smell and a heuristic to address the limitations of the existing rules. We evaluated our heuristic against existing detection rules by manually applying it to 300 unit test cases in Java. Results: Our review identified 56 relevant studies. We found that inadequate interpretations of original definitions of the Eager Test smell led to imprecise detection rules, resulting in a high level of disagreement in detection outcomes. Also, our heuristic detected patterns of eager and non-eager tests that existing rules missed. Conclusion: Our heuristic captures the essence of the Eager Test smell more precisely; hence, it may address practitioners’ concerns regarding the adequacy of existing detection rules
Characterizing and Assessing Test Case and Test Suite Quality
No full textContext: Test cases and test suites (TCS) are central to software testing. High-quality TCS are essential for boosting practitioners’ confidence in testing. However, the quality of a test suite (a collection of test cases) is not merely the sum of the quality of individual test cases, as suite-level factors must also be considered. Achieving high-quality TCS requires defining relevant quality attributes, establishing appropriate measures for their assessment, and determining their importance within different testing contexts. Objective: This thesis aims to (1) provide a consolidated view of TCS quality in terms of quality attributes, quality measures, and context information, (2) determine the relative importance of the quality attributes in practice, and (3) develop a reliable approach for assessing a highly prioritized quality attribute identified by practitioners. Method: We conducted an exploratory study and a tertiary literature review for the first objective, a personal opinion survey for the second, and a comparative experiment with a small-scale evaluation study for the third. Results: We developed a comprehensive TCS quality model grounded in practitioner insights and existing literature. Based on the survey, maintainability emerged as a critical quality attribute where practitioners need further support. A well-known indicator of poor test design that can negatively impact test-case maintainability is the Eager Test smell, which is defined as “when a test method checks several methods of the object to be tested” or “when a test verifies too much functionality.” The results of existing detection tools for eager tests are found to be inconsistent and unreliable. To better support practitioners in assessing test case maintainability, we proposed a novel, unambiguous definition of the Eager Test smell, developed a heuristic to operationalize it, and implemented a detection tool to automate its identification in practice. Our systematic approach in the tertiary review also yielded valuable insights into constructing and validating automated search results using a quasi-gold standard. We generalized these insights into recommendations for enhancing the current search validation approach. Conclusions: The thesis makes three main contributions: (1) at the abstract level, a comprehensive quality model to help practitioners and researchers develop guidelines, templates, or tools for designing new test cases and test suites and assessing existing ones; (2) at the strategic level, identification of contextually important quality attributes; and (3), at the operational level, a refined definition of Eager Test smell, a detection heuristic and a tool prototype implementing the heuristic, advancing maintainability assessment in software testing.
Klimakteriet som en livsfas: kvinnors upplevelser av vårdmötet : Integrerande sammanställning av kvalitativ forskning
No full textBakgrund: Klimakteriet är en livsfas då kvinnors menstruationscykel upphör och är en naturlig del av åldrandet. Upplevelsen kan variera mellan olika kvinnor. Under denna livsfas i kvinnas liv slutar äggstockarna att producera hormonerna östrogen och progesteron. Kvinnor löper större risk att drabbas av depression under klimakteriet, vilket kan leda till andra hälsoproblem som osteoporos och hjärt-kärlsjukdomar. Sjuksköterskan har en avgörande roll i att erbjuda stöd och hjälp för att anpassa sig till förändringar i hälsa och kropp. Roys adaptionsmodell betonar vikten av anpassning till livsförändringar. Sjuksköterskan kan tillämpa modellen för att stödja kvinnors hälsa och välbefinnande under klimakteriet. Vårdmötet mellan kvinnor och hälso- och sjukvårdspersonal är avgörande för att säkerställa högkvalitativ vård och för att förbättra livskvaliteten under klimakteriet. Etablering av en god vårdrelation med förståelse för kvinnors behov är centralt för att uppnå god omvårdnad. Syfte: Syftet med studien var att beskriva kvinnors upplevelser av vårdmötet med hälso- och sjukvården vid klimakterierelaterade hälsoproblem. Metod: En integrerande sammanställning av kvalitativ forskning inspirerad av metasyntes med induktiv ansats. Tolv vetenskapliga artiklar tillämpades och analyserades med Fribergs analysmodell. Resultat: Tre teman framkom i resultatet: nödvändigt med personlig anpassning, avgörande med hälsofrämjande åtgärder och betydelse av information och kunskap. Resultatet ger en inblick i hur kvinnor upplever vårdmötet med hälso- och sjukvården. Slutsats: Kvinnors upplevelser av vårdmötet inkluderade både positiva och negativa upplevelser. Kvinnorna värderade personcentrerad vård och uppvisade behov av ökat stöd för klimakterierelaterade hälsoproblem. Brist på tid och känslan av att inte bli tagen på allvar under vårdmötet försämrade hälsan. Utbildning om egenvård uppskattades och förbättrade välbefinnandet. Hälso- och sjukvårdpersonal bör fortsatt initiera samtal samt erbjuda information om klimakteriet för att öka tillfredsställelsen under denna livsfas
Integrating Stamping Tool Temperature Effects into Early-Stage Process Design : Insights from an Industrial Benchmark
No full textReducing the CO2 footprint has become a key objective in the manufacturing sector, with the automotive industry being no exception. A significant portion of a car body consists of stamped components, making the reduction of CO2 emissions in stamping lines a critical focus. One major contributor to emissions is the high material usage, partially derived from the scrap generation. Additionally, production ramp-ups in critical components, such as side door-inners and wheel housings, often lead to increased defect rates, further exacerbating waste. This work investigates the influence of stamping tool temperature increases during production and its consideration in the early stages of process design. Using a Volvo side door inner as an industrial benchmark, various numerical solutions were explored using AutoForm software. The technical note presents the advantages, limitations, and challenges of these approaches, while highlighting the potential of numerical tools to drive CO2 footprint reduction in stamping processes from an early design stage
Building blocks for large-scale evacuation simulations
No full textEffective evacuation planning is a complex challenge that aims to save lives, reduce travel time, and ensure the provision of essential care to vulnerable individuals. Evacuation simulation models are vital tools for evaluating different scenarios, identifying potential bottlenecks, and optimising strategies for large-scale evacuations.This report contains original and unpublished work that examines how these tools and their components can strengthen disaster preparedness and support decision-makers in managing the complexities of emergency response. It kicks off by outlining the key performance indicators (KPIs) relevant to large-scale evacuation, to then review current simulation approaches and their foundational building blocks. These include functions, algorithms, and models used to simulate and analyse various stages of evacuation: from pre-evacuation processes and individual decision making after a warning is issued, to traffic assignment, road network dynamics, vehicle behaviours,and shelter capacity management.The research work primarily focuses on macroscopic and mesoscopic simulation models, highlighting the strengths and limitations of the different paradigms and frameworks discussed in the literature. The report also features a curated selection of commercially available software, providing a timeline of their development, a comparison of key features, and insights into their real-world application in evacuation planning. This overview is further enriched with a series of case studies illustrating how these tools have been employed in disaster and crisis scenarios around the world.This report is published by Blekinge Institute of Technology (BTH) and is funded by the Swedish Civil Contingencies Agency (Myndigheten för samhällsskydd och beredskap, MSB) as part of the project Digital Decision Support for Large-Scale Evacuation (DISTURB). The DISTURB project is financed through the 2:4 Crisis Preparedness Fund (Anslag 2:4 Krisberedskap), which aims tosupport initiatives that strengthen society’s ability to manage crises and their consequences, and to develop and maintain the capacity for heightened civil defence preparedness.Digital Decision Support for Large-Scale Evacuation (DISTURB
Patienters upplevelse av typ 2-diabetes vid psykisk ohälsa : En allmän litteraturöversikt
No full textBakgrund: Typ 2-diabetes är en kronisk sjukdom som påverkar individens vardag och välbefinnande. Många patienter upplever psykisk ohälsa såsom depression och ångest i samband med sjukdomen, vilket kan komplicera egenvård och försämra livskvalitet. Trots riktlinjer som betonar vikten av att integrera psykosocialt stöd i diabetesvården, upplever många patienter brister i vårdens förmåga att möta dessa behov. Syfte: Syftet var att beskriva patientens upplevelse av att leva med typ 2-diabetes vid psykisk ohälsa. Metod: Studien genomfördes som en allmän litteraturöversikt med en induktiv ansats. Nio vetenskapliga artiklar inkluderades, varav åtta hade kvalitativ metod och en hade mixad metod. Databassökningar genomfördes i CINAHL och PubMed. Samtliga artiklar kvalitetsgranskades och analyserades enligt Fribergs fyra stegsmodell för att identifiera gemensamma teman i resultatet. Resultat: Resultatet presenterades i fyra teman: upplevelser av att diabeteshanteringen överskuggas av psykisk sjukdom, upplevelser av att känslomässiga tillstånd påverkar diabetesegenvård, upplevelser av praktiska utmaningar i diabeteshantering samt upplevelser av diabetesvårdens tillgänglighet och anpassning. Slutsats: Det finns ett tydligt samband mellan psykisk ohälsa och utmaningar i att hantera typ 2-diabetes. En mer individanpassad vård med fokus på psykiskt stöd och känsla av sammanhang kan bidra till förbättrad egenvård och livskvalitet. Sjuksköterskans roll är central i att identifiera och bemöta patientens psykiska och emotionella behov vid typ 2-diabetes
Link analysis through time series decomposition and clustering
No full textWe propose a methodology to identify and analyze similarities in long-term trends of road links using travel speed data. The methodology employs seasonal trend decomposition by LOESS (STL) to extract trend curves from travel speed time series, clearly representing underlying long-term patterns and behavior. These trend curves are then analyzed using the k-means clustering algorithm to group road links based on long-term trends. The resulting clusters offer valuable insights for long-term planning in traffic management, infrastructure development, and identifying potential bottlenecks within the road network. To demonstrate the proposed methodology, we applied it to travel speed data from the European road E4, focusing on the route between Södertälje and Stockholm. The analysis reveals distinct trend characteristics and behaviors, highlighting the diverse nature of traffic patterns in different road links
Patienters upplevelser av psykiska besvär med fokus på oro, ångest och depressiva symtom efter hjärtinfarkt : En litteraturöversikt
No full textBakgrund: En hjärtinfarkt är en oväntad och livshotande händelse som kan leda till fysiska och psykiska konsekvenser. Många patienter upplever känslor av oro, ångest och depressiva symtom såsom nedstämdhet och hopplöshet i efterförloppet. Dessa besvär kan ge en negativ påverkan på återhämtningen och försämra livskvaliteten. För att kunna främja god omvårdnad av hög kvalitet är det avgörande för sjuksköterskan att förstå och bemöta patientens psykiska lidande. Syfte: Syftet med denna litteraturöversikt var att beskriva patienters upplevelser av psykiska besvär med fokus på oro, ångest och depressiva symtom efter genomgången hjärtinfarkt. Metod: Studien är en allmän litteraturöversikt som bygger på befintlig forskning med både kvalitativ och kvantitativ ansats. Data samlades in från databaserna CIINAHL och PubMed. Åtta artiklar valdes ut och analyserades med en induktiv ansats, enligt Fribergs modell. Huvudkategorier identifierades genom en jämförelse av likheter och skillnader i resultaten. Resultat: Fyra huvudkategorier identifierades: oro, ångest, depressiva symtom samt social isolering och ensamhet. Inom dessa kategorier framkom sju underkategorier: livsrelaterad oro, rädsla för återinsjuknande. Kroppsliga reaktioner på ångest, kvarstående ångest. Hopplöshet, nedstämdhet och förlust av livsglädje, brist på motivation och trötthet och sömnstörningar. Slutsats: Efter en hjärtinfarkt är patienters upplevelser av psykiska besvär komplexa och individuella. Oro, ångest och depressiva symtom är vanliga reaktioner som kan påverka vardagen negativt. Genom ökad kunskap om dessa upplevelser kan sjuksköterskor erbjuda individanpassat stöd. För att lindra psykiskt lidande och främja återhämtning är ett personcentrerat förhållningssätt, närvaro och empatiskt bemötande avgörande
Utilization of SAR Backprojection in the Analysis of Downscaled Structures in the D-frequency Band
No full textThe use of THz frequencies for radar remote sensing has provided a new set of applications, where short-range sensing for high-resolution imaging can be used. At the same time, radar imaging of physically large objects for understanding their radar signatures, which is based on long-range remote sensing, can be a cumbersome process and require significant financial resources. In this paper, we propose to explore the synthetic-aperture-radar (SAR) concept at sub-THz frequencies to scale a real-life scenario of long-range imaging of foreign objects down to short-range sensing in the indoor environment. The idea is experimentally studied on four downscaled marine vessels, where the SAR imaging system is based on a D-band FMCW radar that operates at frequencies 126-182 GHz, and the global backprojection algorithm with modified linear interpolator is used for the SAR scene reconstruction. The experimental results demonstrate the feasibility of the proposed idea.