Machine learning is increasingly used to accelerate materials discovery across large compositional and structural design spaces, but limited and heterogeneous data make reliable uncertainty estimation essential. In this work, we investigate uncertainty quantification for permanent magnet modeling across three complementary studies. First, on a public Curie temperature surrogate dataset, we benchmark Gaussian process regression, random forest bagging, and dropout-based Bayesian neural networks and show that calibration, sharpness, and confidence curve diagnostics reveal differences in uncertainty quality that are not visible from point-prediction metrics alone. Second, we transfer this framework to the prediction of intrinsic magnetic properties in Nd2Fe14B-based magnets. Third, we extend the same uncertainty-aware perspective to coercivity prediction from microstructural information using a graph neural network. Together, these studies show that uncertainty quantification improves the trustworthiness of magnetic material property predictions and can be transferred from composition-based surrogate models to more complex structure-sensitive learning tasks.
Freshwater food webs are increasingly exposed to eutrophication and warming, which can shift phytoplankton community composition and reduce the availability of long-chain polyunsaturated fatty acids to higher trophic levels. We investigated whether the model freshwater crustacean Daphnia magna can compensate for a dietary lack of eicosapentaenoic acid (EPA) through endogenous EPA synthesis from its precursors, and whether this response depends on food quality, temperature, and genetic background. Two D. magna strains, isolated from contrasting temperate regions, were exposed to four diets ranging from edible cyanobacteria-dominated to cryptophyte-enriched communities at 15, 20, and 25 °C. Using deuterium labelling, compound-specific hydrogen isotope analysis, fatty acid quantification and fitness measurements, we assessed EPA bioconversion and its consequences for consumer performance. Cyanobacteria-rich food reduced biomass accrual, delayed reproduction, decreased body size, and lowered offspring production. Despite these pronounced fitness effects, endogenous EPA production remained stable across all dietary treatments. Instead, bioconversion was primarily governed by water temperature, reaching a maximum at approximately 20 °C, and was slightly higher in the cold-adapted strain, consistent with underlying biochemical constraints on metabolic processes. Slight differences between the two strains suggest an important contribution of genetic background in determining EPA biosynthesis. These results indicate that endogenous EPA synthesis in D. magna was not actively regulated but was constantly active and potentially buffered against low dietary EPA levels, but did not fully compensate for poor food quality.
Social influences, including group dynamics, social norms and peer support, are assumed to influence physical activity (PA). A previous systematic review and meta-analysis found that 'true groups' (that is, those applying group dynamics principles) had advantages over individual PA interventions. Given technological advances and virtual platforms, we updated previous findings. Systematic searches of electronic databases were completed on 19 March 2024. A meta-analysis of 71 studies (523 effect sizes) compared individual and group-based PA interventions across behavioural, functional (for example, strength and flexibility), psychosocial (for example, quality of life and loneliness) and health outcomes (for example, VO2max). Using a three-level random effects model, group-based conditions showed a small but non-significant advantage for PA (n = 22,042; g = 0.086, 95% CI (-0.061, 0.233), P = 0.249), psychosocial outcomes (n = 18,223; g = 0.292, 95% CI (-0.171, 0.755), P = 0.214) and health outcomes (n = 31,607; g = 0.125, 95% CI (-0.023, 0.272), P = 0.096). For functional outcomes, a significant advantage for group-based interventions emerged after outlier removal (n = 14,429; g = 0.164, 95% CI (0.032, 0.297), P = 0.015). Moderation analysis indicated that different group types produced comparable effects, except that 'true groups' were associated with a larger effect size for health outcomes. Both individual and group-based approaches (in-person or online) may be effective. Future research should explore mechanisms that enhance their impact across populations and settings. This review was prospectively registered with PROSPERO (CRD42021271452).
In digital humanities (DH) and cultural heritage (CH), visualization-based storytelling (VBS) has become an important approach for structuring, interpreting, and communicating cultural data and research results. These domains differ markedly from other VBS application areas through their focus on historical phenomena; their reliance on heterogeneous and semantically ambiguous sources; their engagement with both fictional and factual constellations; their use of qualitative, interpretive, and critical perspectives; and their long-standing expertise in narrative practice and theory. As a result, DH and CH offer particularly rich opportunities for VBS, yet relevant work remains dispersed across humanities scholarship, DH and CH venues, visualization research, public application contexts, and tool-centered communities.Against this backdrop, we survey DH and CH work on story designs and VBS tools to identify trends and recurring patterns, promising practices, and open challenges. We contribute by (i) synthesizing storytelling design spaces into a framework tailored to VBS in DH and CH, (ii) mapping existing approaches to generate a field-level picture of practices and gaps, and (iii) highlighting future areas of concern and inquiry for VBS in relation to domain-specific epistemic questions. Overall, this survey seeks to consolidate an emerging community of practice and provide a shared reference point for future research.
An increasing number of gig workers turn to online gig work platforms to pursue potentially boundaryless and protean careers outside traditional organizational settings. Proactive career self-management and related self-leadership are essential to succeed in such careers. However, while traditional organizations facilitate self-leadership by providing traditional organizational resources, such as supportive human resource practices, gig work lacks these important resources. As a result, cultivating self-leadership among gig workers remains a considerable challenge. This study addresses this challenge by investigating how online platforms that centrally shape work environments may provide a resource-related context that facilitates gig workers' self-leadership strategies. Using a contextual and systemic approach to career self-management and inspired by the conservation of resources theory, we draw on insights from a 14-month field study of an online gig work platform, involving data collected from multiple sources. The findings show how platforms offer various types of platform-embedded resources and how workers utilize these resources to engage in behavior- and cognitive-focused self-leadership strategies. This study thereby contributes to the literature on careers in gig work by theorizing the role of online platforms in shaping gig workers' careers from a proactive and agentic perspective.