Search Results (6,848)

Lightweight intrusion detection systems (IDSs) are increasingly integrated into applied data science workflows for cybersecurity and process monitoring, where limited computational resources and asymmetric error costs constrain model design. This paper presents a comparative study of two lightweight deep learning IDS architectures: ForNet, a convolutional model optimized for feature-centric detection, and SigNet, a gated recurrent model designed for sequence-oriented modeling of ordered flow-feature representations. Both models are trained with Cost-Robust Focal Loss (CRF-Loss), a cost-aware objective that penalizes false positives and false negatives according to deployment-specific risk preferences. We evaluate the models on the UNSW-NB15 and CIC-IDS2017 benchmarks using six standard metrics (accuracy, precision, recall, F1-score, Matthews correlation coefficient (MCC), and the area under the receiver operating characteristic curve (AUROC)), complemented by an analysis of false-positive behavior. On CIC-IDS2017, ForNet achieves precision up to 0.95 and MCC up to 0.93 with AUROC above 0.94, while SigNet shows a stronger recall-oriented profile on UNSW-NB15. In an ablation study, replacing Binary Cross-Entropy with CRF-Loss reduces the false-positive rate by approximately 15–20% and improves robustness-oriented metrics such as MCC by up to 12% on CIC-IDS2017. Rather than claiming universal state-of-the-art performance, the study focuses on performance–risk trade-offs under realistic operational constraints. The results highlight how architectural bias and cost-aware optimisation jointly shape IDS behaviour and offer benchmark-based guidance for interpreting performance–risk trade-offs in lightweight intrusion detection. Full article

While GenAI is transforming education, it remains unclear how it shapes students’ behavior, especially concerning AI literacy. The purpose of this study is to examine which factors positively affect students’ learning behavior and whether AI literacy moderates this effect, using the COM-B model. An online survey of 438 participants was analyzed using covariance-based structural equation modeling (CB-SEM) and fuzzy-set qualitative comparative analysis (fsQCA). The CB-SEM results indicate that independent learning ability, receptive ability, learning environment, AI support equipment, and both intrinsic and extrinsic motivations significantly shape student learning behavior. Notably, AI literacy moderates the relationship between GenAI and learning behavior. Furthermore, fsQCA reveals seven configurations of these factors that favorably impact learning behavior. Together, these findings provide theoretical and practical insights for universities, highlighting actionable ways universities can support students’ adoption of GenAI. Full article

Computational thinking (CT) has become a cross-curriculum priority in many educational jurisdictions, yet a growing body of research reports uneven integration in initial teacher education (ITE), limited preservice teacher confidence, and persistent misconceptions that equate CT with coding. Concurrently, generative artificial intelligence (GenAI) has rapidly entered university programmes, offering new possibilities for modelling problem-solving, generating multiple representations, and supporting iterative design. However, while constructs such as self-efficacy, cognitive load, and affect are well established in educational psychology, their specific application to the intersection of CT and GenAI in teacher education remains under-theorised: existing research has not systematically examined how these psychological dimensions interact when preservice teachers learn CT through GenAI-mediated tasks. This thematic literature review synthesises 54 sources across three intersecting domains: CT frameworks and their pedagogical implications, CT integration in preservice teacher preparation, and GenAI in teacher education and learning design. Drawing on Bandura’s social cognitive theory, cognitive load theory, and research on technology-related affect, the review foregrounds the affective, cognitive, and cultural dimensions of preservice teachers’ engagement with CT and GenAI. The review proposes the GenAI-Enabled Computational Thinking for Preservice Teachers (GECT-P) model, which integrates CT dimensions with GenAI-supported learning cycles, psychological mediators, and teacher education outcomes. The model positions prompting as an epistemic and pedagogical practice that can make CT visible, supports cycles of decomposition, abstraction, pattern recognition, and algorithmic design, and embeds critical AI literacy, ethics, affective scaffolding, and classroom enactment. Design principles and practical pathways are offered for teacher educators seeking to prepare graduates who can develop CT with and beyond GenAI across diverse curriculum areas. Full article

A system of initial teacher education must incorporate instruments capable of capturing the complexity of professional learning. In this context, national diagnostic assessments have become central mechanisms for monitoring outcomes in initial teacher education. This study examines student performance in Initial Teacher Education in Physical Education (ITEPE) programs using the themes assessed by the National Diagnostic Assessment (NDA) 2024, comparing achievement levels and analyzing differences according to the type of higher education institution in Chile. A quantitative, cross-sectional, and comparative design was employed, using official data from 1102 students enrolled in all Chilean universities offering the program. Descriptive and nonparametric inferential analyses were conducted to examine differences by sex and type of institution. The results show relatively homogeneous performance across standards, with higher percentages of achievement in dimensions related to didactic organization and assessment, and lower results in the standard associated with understanding student characteristics. Although statistically significant differences were identified according to administrative dependency, the effect sizes were small. Consequently, the NDA is positioned as a formative input to guide contextualized curricular improvements rather than as a mechanism for institutional ranking. Full article

This study systematizes the form generation process using machine learning-driven motion-tracking data and investigates the interrelationships between the characteristics of generated data and forms generated according to data-processing methods. Through the vision-based machine learning motion estimation (VideoPose3D) algorithm, 3D motion data are extracted from 2D video and categorized into point (joint), curve (bone), and boundary (range of motion) types. Furthermore, this study analyzes the form generation characteristics and limitations associated with each type of motion-tracking data derived from dynamic-to-dynamic physical activities with postural transitions. A data-processing methodology based on artistic practice from prior research is applied. The characteristics of generated data and the morphological characteristics of generated forms are then analyzed according to non-processed and processed methods. Results suggest a potential correlative tendency between the characteristics and generated forms of each type of motion data value information. A bidirectional complementary relationship exists between non-processed and processed motion-tracking data. The data-based form generation methodology demonstrates potential applicability in architectural design. This study expands design possibilities by supporting decisions early in the architectural design process and immediately generating diverse alternatives; it also proposes a standardized framework for a universal data-centric design process applicable to diverse data types, including motion data. Full article

Background/Objectives: Specific Learning Disorders are lifelong neurodevelopmental conditions that persist in adulthood, yet research has traditionally focused on children. In adults, there is significant heterogeneity in cognitive profiles and a lack of consensus on how to operationalize these disorders. This study aims to map the variability in cognitive functioning in university students with Specific Learning Disorders and investigate whether cognitive profiles differ across diagnostic categories and comorbidities. Methods: A retrospective analysis was conducted on the clinical documentation of 166 university students with a diagnosis of Specific Learning Disorders. Participants were categorized into three subgroups: predominant reading disorder, predominant arithmetic disorder, and mixed learning disorder. Cognitive functioning was assessed using Wechsler scales indices. Data were analyzed using linear mixed-effects models and Latent Profile Analysis. Results: Across the sample, reasoning abilities were significantly higher than cognitive efficiency, with working memory consistently emerging as a core weakness. The mixed-disorder group exhibited the lowest cognitive scores and the greatest working memory deficits. Latent Profile Analysis identified two distinct latent subgroups: a “Low Profile” characterized by weaker working memory and a “High Profile” characterized by stronger reasoning and balanced efficiency. Diagnostic labels were only partially aligned with these profiles; while the mixed-disorder group was overrepresented in the “Low Profile,” substantial intra-individual variability existed across all diagnostic categories. Conclusions: The findings suggest that traditional categorical labels for Specific Learning Disorders have limited explanatory power in adulthood, given the high heterogeneity of cognitive functioning. Cognitive weaknesses, particularly in working memory, persist even in high-achieving university students. Clinical and educational support should shift from a label-based approach toward a dimensional, profile-based model to better address the unique strengths and vulnerabilities of adults with Specific Learning Disorders. Full article

There are numerous ways in which generative artificial intelligence (GAI) can be applied in the teaching and learning process. This paper presents one application in the Business Decision Analysis (BDA) course. BDA is considered as the most challenging course in the Graduate Study Program in Economic Entrepreneurship at the University of Zagreb Faculty of Organisation and Informatics; consequently, the teachers continuously analyse possibilities to make the course more attractive for students. The innovative teaching activity at BDA was implemented as a betting shop during the first colloquium (which accounts for 50% of the overall grade). In the activity, GAI analysed learning management system (LMS) data of students’ results (attendance, self-assessment test results, logs in the system) of the initial (pre-course) test, as well as their results of the pub quiz (activity organised a week before the colloquium as a preparatory activity). GAI analysed all the data and predicted the number of points each student will achieve. Additionally, GAI calculated the risk index, average growth (among self-assessment tests) and learning consistency for each student. Finally, GAI created a message for each student that explained what went well in their learning activity, what could be improved, and included a motivational note for the test. The rule was: if a student achieved a higher result than the GAI predicted, the teacher would buy a chocolate for that student. More than 60% percent of students achieved a higher score than was predicted. Surprisingly, exceeding the expected result was not in correlation with the risk indices determined by the GAI. Cluster analysis identified four student profiles consistent with the correlation results, showing weak overall agreement between the predicted and achieved scores, except in the male subgroup, while higher predicted scores were associated with higher average growth and lower risk indices. Qualitative analysis of the GAI application in teaching yielded positive comments, as students perceived the activity as helpful, motivating, and engaging, and would have liked more similar activities. Full article

Network Intrusion Detection Systems (NIDS) face increasing challenges from sophisticated cyber threats, particularly zero-day attacks that evade signature-based methods. While supervised learning is effective for known attack classification, it struggles with novel threats, whereas anomaly-based approaches suffer from high false positive rates and unstable thresholds. To address these limitations, this paper proposes a decision-level adaptive intrusion-detection framework combining hierarchical CNN-based closed-set classification with autoencoder-based zero-day detection in a cascade architecture. The framework enables deployment-time adaptation by dynamically adjusting class-specific confidence thresholds and fusion parameters without model retraining. Experiments on the CSE-CIC-IDS2018 dataset demonstrate strong closed-set performance, achieving 98.98% accuracy and a macro-F1-score of 0.9342, with improved recall for minority attack classes under adaptive thresholding. Under a zero-day evaluation protocol in which Web_Attacks and Infiltration are excluded from training and validation, the proposed approach achieves an F1-score of 0.9319 while maintaining a low false positive rate of 0.0019. The framework is further evaluated on the Simulated University Network Environment (SUNE) dataset representing campus network traffic, achieving 96.18% closed-set accuracy and 97.54% accuracy in the integrated cascade setting. These results demonstrate that the proposed framework effectively balances minority attack detection, zero-day identification, and false-alarm control in dynamic and resource-constrained network environments. Full article

Background/Objectives: Intimate partner abuse (IPA) is common among university students, including nursing students, and is linked to posttraumatic stress symptoms. Accessible online psychosocial interventions are needed to reduce trauma-related symptoms and support posttraumatic growth (PTG). This study examined the effects of an online psychosocial intervention grounded in social learning theory and cognitive behavioral therapy on posttraumatic stress symptoms and PTG among nursing students who experienced IPA in Türkiye. Methods: A randomized controlled trial was conducted among nursing students in Türkiye reporting IPA exposure. Participants were randomly assigned to an intervention group (n = 17) or a control group (n = 18). The intervention group received an eight-session online psychosocial program delivered individually. Assessments were conducted at pre-intervention, post-intervention, and at 1-, 3-, and 6-month follow-ups. Repeated-measures ANOVA was used, and partial eta-squared (ηp²) values were calculated. Results: The intervention group showed significant reductions in posttraumatic stress symptoms compared with the control group, with large effect sizes (p < 0.001; ηp² = 0.402–0.676). Furthermore, significant increases were observed in posttraumatic growth, indicating large and sustained effects over time (p < 0.001; ηp² = 0.515–0.773). Conclusions: The online psychosocial intervention effectively reduced posttraumatic stress symptoms and enhanced posttraumatic growth among nursing students who experienced IPA. However, results should be interpreted with caution due to the small sample size, and future studies with larger samples are warranted. Full article

Distance learners in higher education are often assumed to face limited peer interaction, potentially weakening their sense of belonging. This study examines peer relationships and belonging among students in distance and blended university programs, with attention to the role of virtual reality (VR) within digitally mediated learning environments. Immersive VR teaching is included in the curriculum for distance learning students in the studied programs. Using a mixed-methods design, survey data and open-ended responses were collected from 17 students in Information Studies and Information Service Design. An adapted Classroom Community Scale was supplemented with items addressing the perceived contribution of different communication technologies. Contrary to expectations, fully distance learners did not report weaker agreement with statements reflecting belonging than blended students; on several items, they expressed stronger agreement, particularly regarding perceived peer support and learning opportunities. Results indicate that conventional 2D communication tools, particularly chats and video calls, are central to sustaining peer relationships. VR was not perceived as essential but described by some students as an added value supporting shared experience and group cohesion. Overall, belonging emerges as a socio-technical achievement shaped by communication practices rather than physical proximity. Full article

The undergraduate nursing students’ final year project, commonly called a ‘dissertation’ is an important component of the bachelor’s nursing programme. It can take the form of a literature review and proposal for a research or service improvement project. While crucial for developing research competence and evidence-based practice skills in preparation for their future careers, nursing students often find the dissertation process highly stressful. An online qualitative survey comprising open-ended questions was used to elicit nursing students’ rich, reflective accounts of the dissertation process at a university in the Northeast of England (hereafter referred to as the study site) from those who have recently completed their dissertations. The data obtained from 24 pre-registration nursing students who responded to the survey were thematically analysed. The findings revealed that critical relationships and essential support systems were key mediators of the challenges students faced, particularly a lack of readiness for the dissertation module, but they ultimately achieved transformative outcomes of an effective learning experience. Their navigational challenges can inform curriculum design and practices to better support students in their dissertation journey. Full article

Positioned at the intersection of instructional mediation, Visual Education, and Universal Design for Learning (UDL), this research aims to ascertain whether the use of Artificial Intelligence (AI) enhances accessibility for students with sensory disabilities. The study involved 137 pre-service teachers attending the “Special Didactics and Learning for Sensory Disabilities” course within the teacher specialization program (TFA) at the University of Foggia. Although the hybridization of AI, UDL, and Visual Education was favourably received, its application remains sporadic, highlighting the challenge of balancing the need for simplification with requisite conceptual accuracy. This underscores the necessity of integrating more structured and continuous training pathways into teacher education, grounded in visual education and featuring micro-modules dedicated to specific skills such as writing alternative text, subtitling, and verifying color contrast according to recognized standards. Full article

As part of a NASA University Leadership Initiative (ULI) program, this work supports the continued development and evaluation of a fatigue-based process window for stress-relieved Ti-6Al-4V specimens produced via laser powder bed fusion (PBF-LB/M). Four-point bend and axial fatigue specimens were fabricated by NASA ULI collaborators across a range of scan velocities (800–2000 mm/s) at a constant power of 370 W using an EOS M290 system. All fatigue specimens were low-stress-ground by a commercial vendor and tested at Case Western Reserve University (CWRU) under load-controlled cyclic loading at a stress ratio of R = 0.1. This paper presents a curated dataset linking PBF-LB/M process parameters to fatigue outcomes across 175 specimens. Of these, 136 fractured and this study includes fatigue crack initiation site identification and defect morphology metrics derived from post mortem SEM analysis. Specimens that reached runout (10⁷ cycles) and did not fracture under subsequent fatigue testing are retained in the dataset, with fractographic fields marked as ‘NA’ to indicate non-applicability. The dataset includes specimen metadata, processing parameters, fatigue life data, fatigue initiation site classification (e.g., keyhole, gas-entrapped pore (GeP), lack-of-fusion (LoF), contamination), defect size and shape descriptors, and spatial location relative to the free surface. These data are intended to support defect-based fatigue life prediction, probabilistic modeling, process–structure–property studies, and machine learning frameworks linking process parameters to fatigue performance in PBF-LB/M Ti-6Al-4V. Full article

Residential college (RC) activities represent a creative form of engagement beyond formal classroom teaching, integrating experiential, social, and community-based learning. China is promoting holistic education through the RC activities. Student motivation directly influences their engagement in practice-based learning. Yet, the motivation profile of Chinese students’ participation in RC activities remains largely unexplored. To address this gap, the exploratory cross-sectional study involved 403 undergraduates from an RC-based university in the Guangdong–Hong Kong–Macao Greater Bay Area. Based on a standardized questionnaire, we explored descriptive patterns of three forms of motivation, namely intrinsic motivation, extrinsic motivation, and amotivation, as well as four influencing factors: activity logistics, experiential value, outcome expectations, and social context. We then conducted hierarchical linear regression analyses with the three motivation types as outcomes. The primary results indicated that intrinsic motivation was the dominant motivation type among RC students. Experiential value emerged as a key influencing factor, positively associated with intrinsic and extrinsic motivation, and negatively associated with amotivation. Additionally, activity hosting experience was identified as another important correlate, positively linked to both intrinsic and extrinsic motivation. Motivational patterns further varied across gender, academic year, ID place, weekly RC stay duration, and part-time employment. The findings provide empirical support for more targeted RC activity planning aimed at boosting student motivation in the Chinese context. Full article

This study seeks to examine the attractiveness of higher education institutions as community spaces for students and the significance of the services they provide. It aims to explore students’ perspectives on planning their long-term futures, particularly in assessing whether the university environment serves as a stronger influence than their place of origin or family background. The role of the university is particularly significant if it is located outside the student’s town of origin, so the student’s decision to attend a particular institution is not necessarily based on the specific undergraduate program or the prestige of the university. The study combines a review of the national and international literature with an empirical investigation, utilizing a questionnaire survey to analyze students’ decision-making processes. Many students perceived the university as a transitional “island”, offering a temporary space to inhabit before embarking on their future careers. The degree obtained serves as a “passport” to professional opportunities, while the university experience provides a unique community environment and represents a significant step toward independence and separation from familial influence. These findings hold particular relevance for universities, which are continually redefining their roles in response to changing student expectations. Many students view the university not merely as a site of learning but as a precursor to adulthood and a foundational space for personal growth. This study addresses a gap in the existing literature by focusing on the appeal of universities as local hubs and comparing their influence to the retaining power of family ties, offering insights for student development. Full article