Open research Europe最新文献

{"title":"The pursuit of equity in COVID-19 policy and policymaking: A qualitative systematic review.","authors":"Paul Cairney, Stefano Intropido","doi":"10.12688/openreseurope.20563.2","DOIUrl":"10.12688/openreseurope.20563.2","url":null,"abstract":"<p><strong>Background: </strong>The COVID-19 pandemic produced a devastating and unequal effect on global population health and wellbeing. Although research demonstrated multiple COVID-19 inequalities, the pursuit of equity (to address unfair inequalities) remained politically contested and overshadowed by higher priority crisis responses. It is essential to learn from these experiences to inform future crisis responses and anticipate the lack of proportionate and sustained attention to inequalities. We seek to understand how COVID-19 equity research defines this policy problem, offers solutions, and considers their feasibility in complex and political policy processes.</p><p><strong>Methods: </strong>We conducted a qualitative systematic review (2024) to identify peer reviewed journal articles on COVID-19, policymaking, and equity in three databases (Web of Science, Scopus, Proquest). We sought articles providing a non-trivial reference to policymaking concepts, including 55 texts that meet the inclusion criteria, and adding 30 texts by snowballing. We used an immersive and inductive approach to identify key themes and show how the use of policy concepts and theories informs an overall narrative of COVID-19 equity research.</p><p><strong>Results: </strong>This research documents the unequal impact of the COVID-19 pandemic and policy, then identifies potential policy solutions and some hopes that governments will support them. However, it highlights a major gap between this aspiration for change versus political reality, and identifies barriers to the production and use of lessons for future crises.</p><p><strong>Conclusion: </strong>Most governments appear to have learned few lessons about inequalities from their COVID-19 experiences. The literature suggests that most governments will contribute to very similar unfair inequalities in their responses to the next crisis.</p>","PeriodicalId":74359,"journal":{"name":"Open research Europe","volume":"5 ","pages":"167"},"PeriodicalIF":0.0,"publicationDate":"2026-06-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13324367/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148392648","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Unlocking regional bioeconomy transitions: Co-creation, stakeholder engagement, and systemic change in rural Europe.","authors":"Margit Hofer, Martina Lindorfer, Samire Gurgurovci","doi":"10.12688/openreseurope.21341.2","DOIUrl":"10.12688/openreseurope.21341.2","url":null,"abstract":"<p><strong>Background: </strong>Transitioning to a sustainable bioeconomy in rural Europe is challenged by the difficulty of translating high-level strategies into locally relevant initiatives. Regional bioeconomy hubs aim to bridge this gap, but their success hinges on engaging diverse, often conservative, stakeholders. While co-creation is a promising approach, its practical application in these contexts is poorly understood. This study examines the establishment of nine hubs within the Horizon Europe project RuralBioUp to identify effective strategies for fostering stakeholder ownership and systemic change.</p><p><strong>Methods: </strong>This qualitative study analysed multi-layered data collected from nine regional bioeconomy hubs between October 2022 and June 2025. Sources included seven semi-structured interviews with hub facilitators, two mutual learning workshops, participatory feedback tools, and co-created documentation. Thematic analysis identified patterns related to effective actions, co-creation dynamics, and innovation adoption.</p><p><strong>Results: </strong>Effective engagement actions were practical and locally grounded, such as study visits, targeted training in local languages, and co-organised events. Generic online workshops and bureaucratic activities proved ineffective. Co-creation was enabled by institutional flexibility, trusted local facilitators, and established networks. Key barriers included stakeholder time scarcity, logistical challenges, cultural scepticism, and unclear value propositions. In conservative settings, innovation adoption was facilitated through peer-to-peer learning and framing new ideas as extensions of existing practices.</p><p><strong>Conclusions: </strong>The success of regional bioeconomy hubs is contingent on trust, perceived local relevance, and stakeholder ownership, not just technological novelty. Instead of one-size-fits-all models, patient, bottom-up strategies that co-adapt innovation to community values are essential for long-term change. Hubs can act as critical facilitators for place-based transitions by embedding co-creation within flexible governance structures. Policy and practice should prioritize context-sensitive, tangible engagement to foster inclusive and sustainable rural bioeconomy development.</p>","PeriodicalId":74359,"journal":{"name":"Open research Europe","volume":"5 ","pages":"392"},"PeriodicalIF":0.0,"publicationDate":"2026-06-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13550885/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148922045","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Training resources on open and responsible research and innovation (Open RRI): gaps, quality and opportunities to advance training.","authors":"Cristina Lagido, Kristian Nielsen, Gitte Kragh, Pascal Flohr, Louise Bezuidenhout, Loek Brinkman, Agnese Baini, Chiara Saviane, Asya Salnikova, Adam Brandstetter-Kunc, Giacomo Consolini, Antónia Correia, Paula Moura, Maya Fedeli, Jonathan England, Shanmugasundaram Venkataraman, Aslak Skarvøy, Teodora Konach, Borana Taraj, Katharina Koller, Ilse Marschalek, Maria Schrammel, Sanja Jurkovic, Bojan Macan, Alen Vodopijevec, Brian Cahill, Andrea Giraldo Sevilla, Nico Pitrelli, Alessio Livio Spera, Claudia Iasillo, Pedro Príncipe, Noemi De Lorenzo, Muki Haklay","doi":"10.12688/openreseurope.23510.1","DOIUrl":"10.12688/openreseurope.23510.1","url":null,"abstract":"<p><strong>Background: </strong>Open Science (OS) and Responsible Research and Innovation (RRI) are increasingly converging under the concept of Open RRI, which integrates openness with inclusivity, responsibility, and societal engagement. Strengthening researcher skills in this area requires a clear understanding of existing training provision. This study presents findings from the Horizon Europe project PATTERN, which mapped and assessed Open RRI training resources across Europe.</p><p><strong>Methods: </strong>A mixed-methods approach was applied, combining desk research, surveys, semi-structured interviews, data workshops, and mutual learning events. In total, 571 training resources were identified and analysed across eight skill areas: Open Access; FAIR Data; Citizen Science; Research Integrity; Gender, Non-discrimination, and Inclusion; Dissemination and Exploitation; Science Communication; and Management and Leadership. A subset of resources was further evaluated using defined quality criteria.</p><p><strong>Results: </strong>The mapping reveals a fragmented and uneven training landscape, with a strong predominance of introductory-level resources and limited provision for intermediate and advanced learners. Training is often generic and lacks contextualisation, with gaps in inclusivity, multilingual provision, and local adaptation. Quality assessment highlights strengths such as modular design, open licensing, and the use of active learning approaches. However, recurring challenges include limited accessibility, unclear pathways for reuse, and difficulties in implementation and long-term sustainability. Key thematic gaps include intersectionality, policy engagement, supervision ethics, and the integration of indigenous and local knowledge.</p><p><strong>Conclusions: </strong>Despite growing availability of training resources, significant gaps remain in scope, depth, and inclusivity. Addressing these challenges requires improved coordination, expansion of advanced and context-sensitive training, enhanced multilingual accessibility, and stronger integration into curricula and institutional practices to support sustainable Open RRI capacity building.</p>","PeriodicalId":74359,"journal":{"name":"Open research Europe","volume":"6 ","pages":"217"},"PeriodicalIF":0.0,"publicationDate":"2026-06-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13560565/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148957457","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Quaternary aminostratigraphies for the eastern North European Plain.","authors":"Ellie Nelson, Dustin White, Lucy Wheeler, Stefan Meng, Marcin Szymanek, Jaqueline Strahl, Michael Hein, Witold P Alexandrowicz, Brigitte Urban, Samantha Greeves, Mareike Stahlschmidt, Ralf-Dietrich Kahlke, Tobias Lauer, David Colin Tanner, Kirsty E H Penkman","doi":"10.12688/openreseurope.21815.2","DOIUrl":"10.12688/openreseurope.21815.2","url":null,"abstract":"<p><p>The eastern North European Plain is an important area for studying Quaternary climate change and archaeology; however, providing chronological constraints for sedimentary deposits can be challenging. Amino acid geochronology (AAG) is a relative dating technique that has been useful in correlating isolated Quaternary sequences. The intra-crystalline protein decomposition (IcPD) approach to AAG using the opercula of <i>Bithynia</i> snails has previously been used to provide relative dating frameworks across northern and central Europe in areas where the integrated diagenetic temperature can be assumed to be similar. Here, the first aminostratigraphies for the eastern North European Plain are presented, incorporating deposits from at least the last ~1 Ma, which are used to assess the current age attributions to Middle and Late Pleistocene interglacials. These aminostratigraphies are then used to explore expected differences in the extent of IcPD due to differing temperature histories across the study area. Correlations of opercula to regional pollen assemblages representative of the Holsteinian, Eemian and Holocene are used to evaluate the temporal resolution achievable by IcPD within a given interglacial. This work has produced four new aminostratigraphies that can now be used as reference datasets for relative age estimation for the late Middle Pleistocene to the Holocene in the eastern North European Plain.</p>","PeriodicalId":74359,"journal":{"name":"Open research Europe","volume":"5 ","pages":"396"},"PeriodicalIF":0.0,"publicationDate":"2026-06-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12811728/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145999781","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Humanitarian data infrastructures for missing migrants: A multimodal and ethics-integrated framework.","authors":"M Fevzi Esen, Ecenur Aydemir","doi":"10.12688/openreseurope.23124.2","DOIUrl":"10.12688/openreseurope.23124.2","url":null,"abstract":"<p><p>Migrant disappearances have become a critical humanitarian crisis on global scale. However, the lack of strong, consolidated data across irregular migration routes is a barrier to effective intervention. In this paper, we aim to employ a sociotechnical systems perspective to design an ethics-infused infrastructure. Our primary objective is to convert multimodal digital traces into actionable intelligence, thereby supporting search, identification, and policy accountability. To standardize the analysis of decentralized digital evidence, we define a typology of 32 trace types encompassing ten domains, categorized from geospatial markers to psychosocial signals. This classification is the basis for the modular architecture of five linked cores. We present a novel approach that combines entity resolution and spatiotemporal reconstruction with provenance tracking and explainable AI to support effective humanitarian action. These features enable robust early warning capabilities and provide decision support aligned with OCHA standards. The framework aims to facilitate compliance with FAIR data principles and international human rights standards through embedding cross-layer safeguards including dynamic consent, auditability and interoperability. This research provides a scalable, replicable method to validate digital evidence throughout the migration cycle. It bridges computational innovation and rights-based governance. Future work should focus on operational pilots to assess feasibility, triage efficiency, identification accuracy, and family notification protocols.</p>","PeriodicalId":74359,"journal":{"name":"Open research Europe","volume":"6 ","pages":"118"},"PeriodicalIF":0.0,"publicationDate":"2026-06-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13332383/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148392765","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Applying advanced psychometrics to antipsychotic treatment trials: A protocol comparing effect estimates.","authors":"Timo Schurr, Fiona Boland, David Byrne, Frank Doyle","doi":"10.12688/openreseurope.23623.1","DOIUrl":"10.12688/openreseurope.23623.1","url":null,"abstract":"<p><strong>Background: </strong>Antipsychotic trials in schizophrenia quantify treatment efficacy using clinician-rated scales such as the Positive and Negative Syndrome Scale (PANSS) and Brief Psychiatric Rating Scale (BPRS). If these scales include redundant or weakly informative items, measurement noise can blur drug-comparator differences that may dilute real effect sizes. Modern psychometric methods can refine and re-weight symptom scores that may improve precision and potentially increase detectable treatment effects, although it is unclear which approaches might be best.</p><p><strong>Aims: </strong>To 1) determine best psychometric models for the PANSS and BPRS, and 2) test whether psychometrically-informed scoring of PANSS/BPRS alters estimated antipsychotic effect sizes compared with conventional total-score analyses, and 3) to explore whether any changes differ across symptom domains and by sex.</p><p><strong>Method: </strong>We will conduct secondary analyses of anonymised individual participant data from randomized Phase II-IV schizophrenia trials accessed via secure data-sharing platforms (YODA Project and Vivli). Item-level PANSS/BPRS data will be harmonised at baseline and outcome. Psychometric models will be derived using confirmatory factor analysis, item response theory, and network analysis. Stability and invariance analyses will be conducted, examining covariates such as age and sex. Antipsychotic efficacy will be estimated using mixed-effects participant-level models, expressed as Cohen's d standardised mean differences (SMDs) for both conventional and psychometrically-informed outcomes. The primary outcome will be the difference in SMDs between conventional and psychometrically-informed effect sizes, summarised across trials.</p><p><strong>Conclusions: </strong>This study will determine whether best practice psychometric refinement of existing measures meaningfully changes antipsychotic effect size estimates, informing outcome measurement choices and interpretation of trial results.</p>","PeriodicalId":74359,"journal":{"name":"Open research Europe","volume":"6 ","pages":"211"},"PeriodicalIF":0.0,"publicationDate":"2026-06-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13560577/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148957563","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"New challenges in horticultural IPM: artificial intelligence can enhance the use of biocontrol agents.","authors":"Andrea Lovazzano","doi":"10.12688/openreseurope.23884.1","DOIUrl":"10.12688/openreseurope.23884.1","url":null,"abstract":"<p><p>Modern horticulture faces the dual imperatives of increasing global food production while mitigating the severe environmental and health impacts of conventional chemical pesticides. Biological control agents (BCAs), which utilize beneficial microorganisms, macroorganisms, semiochemical and botanicals to suppress pathogens, represent a cornerstone of sustainable agriculture but have been historically constrained by inconsistent field performance, high specificity, and slow action. Concurrently, Artificial Intelligence (AI) is emerging as a transformative force in agriculture, offering a powerful suite of tools for data analysis, prediction, and automation. The synergy of BCAs and AI creates advanced strategies where AI directly addresses the inherent limitations of biological control. It can foster the deployment of AI-driven predictive models for proactive pest outbreak forecasting, enabling timely and effective BCA application. Furthermore, it details the role of precision robotics and drones, guided by computer vision, in the targeted delivery of these agents. These components are synthesized into the concept of Integrated Pest Management (IPM) 5.0, where intelligent decision support systems orchestrate a holistic, data-driven approach to plant health. However, the realization of this vision is contingent on overcoming significant economic, regulatory, and adoption hurdles. High initial costs, complex and divergent regulatory countries in the EU and US, and socio-technical barriers to farmer adoption present formidable challenges. A focus on developing robust, low-cost technologies, enhancing BCA formulation and stability, creating interoperable data frameworks, and addressing the socio-economic factors is necessary to translate technological potential into widespread, sustainable practice.</p>","PeriodicalId":74359,"journal":{"name":"Open research Europe","volume":"6 ","pages":"204"},"PeriodicalIF":0.0,"publicationDate":"2026-06-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13679776/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149231356","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A conversational AI toolkit for citizen deliberation in party primary elections.","authors":"F Ramón Villaplana, Michal Malý","doi":"10.12688/openreseurope.24102.1","DOIUrl":"https://doi.org/10.12688/openreseurope.24102.1","url":null,"abstract":"<p><p>Primary elections within political parties are increasingly consequential for democratic outcomes, yet voters -both party members and the general public- frequently lack systematic, comparable and trustworthy information about the candidates competing for nomination. Traditional voter information tools, such as Voting Advice Applications, focus on programmatic alignment in general elections and rarely address the specific demands of intraparty contests, where electability, leadership style, integrity, and organisational capital matter alongside policy positions. The recent availability of large language models with web-search capabilities offers an opportunity to develop accessible, adaptable and free analytical tools that systematise public information about candidates while preserving voter autonomy. We developed PRIMARYCAN, an open-source prompt-based toolkit consisting of three components: (1) a main analytical prompt that generates structured candidate profiles in a default accessible mode and an optional professional mode; (2) an external evaluator prompt for auditing outputs in a different model; and (3) a methodological guide. The toolkit produces structured analyses that are comparable across candidates, traceable to public sources, and adaptable to user expertise. Six analytical blocks cover identity and trajectory, discourse and communication, capabilities and performance, consistency and integrity, political capital and electoral potential and (optionally) advanced behavioural indicators. PRIMARYCAN includes safeguards against common pitfalls in algorithmic political analysis: voting verdicts are prohibited; criteria symmetry is enforced; behavioural patterns replace clinical psychometric labels; hindsight bias safeguards are mandatory in retrospective analyses. The toolkit illustrates that conversational AI, when carefully scaffolded with explicit ethical and epistemic constraints, can serve as a useful intermediary for democratic deliberation in primary elections. We discuss limitations (model variability, hallucination risk, structural limits of self-evaluation), pending empirical validation with real users, and pathways for adaptation to other electoral contexts.</p>","PeriodicalId":74359,"journal":{"name":"Open research Europe","volume":"6 ","pages":"201"},"PeriodicalIF":0.0,"publicationDate":"2026-06-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13562963/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148957480","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"AI-based services for inclusive language learning in immersive XR environments: Speech translation, and sign language integration.","authors":"Nikolaos D Tantaroudas, Andrew J McCracken, Ilias Karachalios, Evangelos Papatheou","doi":"10.12688/openreseurope.23214.2","DOIUrl":"10.12688/openreseurope.23214.2","url":null,"abstract":"<p><strong>Background: </strong>Extended Reality (XR) technologies offer transformative potential for language education, yet current platforms largely neglect the accessibility needs of deaf and hard-of-hearing individuals. Existing solutions typically operate in single-language environments and lack integrated support for sign language and multimodal communication. There is a critical need for inclusive platforms that serve both deaf and hearing learners through cross-modal AI services embedded in immersive environments.</p><p><strong>Methods: </strong>This study presents a modular platform integrating six AI services: speech-to-text transcription (OpenAI Whisper), multilingual translation (Meta NLLB), text-to-speech synthesis (AWS Polly), sentiment analysis (RoBERTa), session summarisation (flan-t5-base-samsum), and International Sign (IS) translation via Google MediaPipe. An IS dataset of 750 gesture videos was processed to extract hand landmark coordinates mapped to 3D avatar animations within a Unity-based VR environment on Meta Quest 3 headsets. The system was validated through technical benchmarking of AI service performance, including comparative evaluation of text-to-speech services and multilingual translation models (NLLB-200 and EuroLLM 1.7B variants), load testing to assess platform. scalability, and end-to-end pipeline latency measurement for both the hearing and the deaf user pathways. The educational scenario was additionally evaluated in a companion pilot study, ⁵⁰ which shares the same underlying AI services and provides complementary user-perception evidence.</p><p><strong>Results: </strong>Technical benchmarking confirmed the platform's viability for real-time XR deployment. TTS benchmarking confirmed AWS Polly's lowest latency (50-100 ms first byte) at competitive cost. The EuroLLM 1.7B Instruct model achieved a BLEU score of 84.34, outperforming NLLB's 79.25. Load testing with 1,000 simulated concurrent users demonstrated average response times under 800 milliseconds with no critical failures. Avatar animation latency for IS sign rendering remained consistently under 300 milliseconds. End-to-end pipeline latency averaged 2.05 ± 0.31 s for the hearing pathway and 2.32 ± 0.34 s for the deaf (IS) pathway, both within accepted thresholds for conversational educational use. The companion pilot (N = 10) reported a mean overall experience rating of 4.6/5.0, 92% user satisfaction and unanimous (100%) demand for expanded language and sign-language support. ⁵⁰.</p><p><strong>Conclusions: </strong>The results presented in this study focus on the technical feasibility of integrating cross-modal AI services within XR environments for accessible, multilingual language learning. The modular architecture enables independent scaling and adaptation to diverse contexts, laying the groundwork for equitable educational solutions aligned with EU digital accessibility objectives.</p>","PeriodicalId":74359,"journal":{"name":"Open research Europe","volume":"6 ","pages":"72"},"PeriodicalIF":0.0,"publicationDate":"2026-06-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13324358/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148392807","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"SentinelSphere: An AI-driven cybersecurity platform integrating real-time threat detection with security awareness education.","authors":"Nikolaos D Tantaroudas, Ilias Karachalios, Andrew J McCracken","doi":"10.12688/openreseurope.22957.2","DOIUrl":"10.12688/openreseurope.22957.2","url":null,"abstract":"<p><strong>Background: </strong>The cybersecurity domain faces dual challenges: a global shortage of qualified professionals and persistent human-factor vulnerabilities contributing to the majority of security breaches. Traditional Security Information and Event Management (SIEM) systems generate excessive false positives causing analyst alert fatigue, while conventional awareness programmes demonstrate limited effectiveness in changing user behaviour. Integrated solutions addressing both technical detection and human-centric education are needed.</p><p><strong>Methods: </strong>We developed SentinelSphere, an AI-driven platform combining machine learning-based threat detection with Large Language Model (LLM)-powered security training. The detection component employs an Enhanced Deep Neural Network trained on the CIC-IDS2017 and CIC-DDoS2019 benchmark datasets. To mitigate the substantial class imbalance present in these datasets (e.g. 168,186 benign samples versus 21 SQL injection samples), class weights inversely proportional to class frequencies were applied during training. The educational component utilises Microsoft's Phi-4 model (14 billion parameters) with quantisation techniques to enable deployment on standard hardware. System performance was optimised through Rust-based preprocessing and validated via pilot deployments across two workshops with a total of 79 stakeholders (3 professionals in maritime, educational and manufacturing sectors; 76 in an educational setting) in Greece.</p><p><strong>Results: </strong>The Enhanced DNN achieved high detection accuracy with significant false positive reduction compared to baseline models, while maintaining strong recall for critical attack categories including DDoS, botnet, and brute force attacks. Rust optimisation delivered substantial speedup in both single-record and batch processing. Stakeholder validation revealed high satisfaction rates, with most participants achieving improved security concept comprehension post-training. The platform identified critical awareness gaps, particularly regarding data protection regulations and multi-factor authentication adoption.</p><p><strong>Conclusions: </strong>SentinelSphere demonstrates that integrating intelligent threat detection with adaptive, LLM-powered security education shows strong potential to address both technical detection and human security awareness challenges, although longitudinal studies are required to validate sustained behavioural change. The resource-efficient design enables deployment in SME environments without enterprise-grade infrastructure, supporting comprehensive cyber resilience approaches within the European Union's regulatory framework.</p>","PeriodicalId":74359,"journal":{"name":"Open research Europe","volume":"6 ","pages":"58"},"PeriodicalIF":0.0,"publicationDate":"2026-06-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13276484/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148297195","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}