RUN
Repositório Institucional da Universidade NOVA de Lisboa
Entradas recentes
Circular economy modelling for climate change mitigation
Publication . Zanon-Zotin, Marianne; Fortes, Patricia; Corbier, Darius; Deetman, Sebastiaan; Edelenbosch, Oreane; Engelenburg, Martijn van; Hauenstein, Christian; Hertwich, Edgar; Jiang, Meng; Köckritz, Luja von; Magalar, Leticia; Pauliuk, Stefan; Straub, Lucas; Vélez-Henao, Johan; Vuuren, Detlef van; CENSE - Centro de Investigação em Ambiente e Sustentabilidade; Elsevier
Circular economy (CE) measures can play a role in reducing greenhouse gas (GHG) emissions, especially in material- and energy-intensive sectors. Yet, their representation in GHG mitigation pathways is still not well captured in climate change mitigation models. To support modelling efforts towards better CE representation, two reviews were conducted: (1) the empirical evidence on the GHG mitigation potential of CE strategies across material supply and demand sectors and R-strategies, and (2) the current modelling approaches used to assess these strategies. Findings show that, while most studies focus on recycling, increasing attention is given to upstream strategies such as material substitution, design for reuse, and service-based business models. Important gaps remain, particularly around Refuse, Rethink, behavioural factors, rebound effects, as well as synergies and trade-offs with climate policy. Industrial ecology methods provide a detailed material flows representation but lack feedback mechanisms and economic dynamics. In contrast, GHG mitigation models offer broader system coverage but often simplify materials and CE dynamics. Better alignment between methods is needed, including shared definitions, improved data, and more collaboration across modelling communities. Strengthening the modelling of CE strategies can support policy-relevant assessments of CE’s contribution to achieving global climate goals.
An Activity-Centric Intelligent Information System for Context-Aware Recommendation in Exploratory Journalistic Research
Publication . Neves-Silva, Rui; Pina, Paulo; UNINOVA-Instituto de Desenvolvimento de Novas Tecnologias; CTS - Centro de Tecnologia e Sistemas; MDPI - Multidisciplinary Digital Publishing Institute
Journalists navigating large digital archives face a tension between speed and thoroughness. Keyword search often returns unmanageable volumes of results, while topic-only recommendation may fail to identify contextually appropriate content. This paper presents magknet, a recommendation system that surfaces archival content for journalists by combining topic classification with hierarchical context modelling, with an explicit focus on transparency and explainability. The system integrates topic-based classification using Vector Space Models over predefined fundamental topics with context-based filtering through a four-dimensional ontological model representing where, when, who, and what. Recommendations are generated primarily from interpretable topic and context similarity, while the architecture allows later reinforcement with observed user attachment patterns. The system was validated at TRL 7 through exploratory testing with professional journalists. All assessed recommendations met the predefined relevance threshold and were judged contextually appropriate within the tested scenarios. Perceived value was strong: all participants recognized value in incorporating the system into professional work, while 84.6% valued the system as an information aggregator and anticipated productivity gains. The system’s value as a memory record was lower (38.5%), suggesting that active recommendation is the stronger use case. Qualitative feedback showed appreciation for contextual relevance and transparency, alongside challenges related to workflow integration and real-time information needs. The paper contributes: (i) an activity-centric formulation of recommendation for professional knowledge work; (ii) a hybrid topic-context ranking model combining vector-space topic representation with ontology-based contextual similarity; and (iii) an exploratory validation showing perceived usefulness, contextual adequacy, and adoption constraints. The results support interpretable, context-aware recommendation for archival journalistic research, while indicating that future systems should be embedded in existing professional tools.
Cost-Effectiveness Analysis of Chatbot-Supported Remote Patient Monitoring for Anticoagulation Management
Publication . Santos, Ana Rita; Sampaio, Filipa; Guede-Fernández, Federico; Perelman, Julian; Londral, Ana; Escola Nacional de Saúde Pública (ENSP); Comprehensive Health Research Centre (CHRC) - Pólo ENSP; LIBPhys-UNL; DF – Departamento de Física; Comprehensive Health Research Centre (CHRC) - pólo NMS; JMIR Publications
Background: Digital health technologies (DHTs) are increasingly integrated into clinical practice, yet economic evaluations remain scarce, particularly in the early development stages. Within the NICE (National Institute for Health and Care Excellence) Evidence Standards Framework, Tier C DHTs comprise technologies with direct clinical implications and measurable health outcomes, for which robust economic evidence is essential. Early-stage assessments are particularly important to inform subsequent development, refinement, and adoption decisions across the digital health lifecycle. Objective: This study aims to explore the feasibility of integrating a full trial-based economic evaluation within an early-stage pilot comparing a chatbot-supported remote patient monitoring (RPM) solution for anticoagulation management with the standard of care (SOC). Methods: A cost-effectiveness analysis was performed alongside a pilot crossover trial among adult cardiac surgery patients receiving vitamin K antagonists. Participants were allocated to two 6-month sequences (SOC→RPM or RPM→SOC). The intervention consisted of a rule-based chatbot integrated with home-based international normalized ratio self-testing using portable coagulometers to support communication and therapy management. Effectiveness was measured as time in therapeutic range (TTR), and costs were estimated from the Portuguese National Health Service and a limited societal perspective over a 1-year horizon. The analysis 1 applied a within-patient cost-effectiveness approach to estimate incremental costs, incremental TTR, and incremental cost-effectiveness ratios. Uncertainty was explored through nonparametric bootstrapping (5000 replications) and deterministic sensitivity analyses. Complementary comparisons examined differences between sequences (analysis 2), between periods (analysis 3), and within each sequence (analysis 4). Results: A total of 19 patients were included in the analyses. In analysis 1, RPM improved anticoagulation control, with a mean within-patient increase of 10.43 percentage points in time in TTR. The mean incremental costs were €198.61 (€1=US $1.08) from the Serviço Nacional de Saúde perspective and €270.05 from the limited societal perspective. The corresponding incremental cost-effectiveness ratios were €19.03 and €25.88 per additional percentage point of TTR gained. Sensitivity analyses produced consistent estimates across parameter variations. Complementary analyses (2-4) suggested that RPM tended to be more cost-effective when implemented after the initial 6-month postoperative period. Conclusions: This proof-of-concept study demonstrates that a full trial-based economic evaluation can feasibly be embedded within an early-stage Tier C DHT. The intervention showed improved anticoagulation control alongside higher costs, providing initial insights into its cost-effectiveness profile. Positioned within the digital health evidence continuum, such assessments can function as a learning stage within the lifecycle. To address the persistent adoption-evidence gap, tier- and stage-aligned frameworks are needed to guide the economic evaluation of DHTs. This study contributes to that goal by providing a set of recommendations specifically for Tier C DHTs.
The Hidden Side of Digital Inclusion
Publication . Macedo, Joana; Neves, Joana; Neves, Catarina; Oliveira, Tiago; Cruz-Jesus, Frederico; NOVA Information Management School (NOVA IMS); Information Management Research Center (MagIC) - NOVA Information Management School
Digital inclusion extends beyond access to platforms—it requires individuals to feel competent and autonomous in online spaces. While much research focuses on how social media affects mental health, this study examines the inverse: how poor mental health shapes experiences of digital participation. Drawing on self-determination theory, we analyze the effects of sleep quality, anxiety, and depression on perceived social media competence, autonomy, and relatedness. Results from a structural equation model show that anxiety and depression reduce competence and autonomy, while poor sleep undermines relatedness. Age moderates the link between relatedness and competence, with older users reporting lower perceived competence. These findings suggest that internal psychological states act as hidden barriers to digital inclusion, especially across age groups. The study contributes to conversations on identity and inclusion by highlighting how mental health influences one’s ability to engage meaningfully in social media environments.
Predicting student churn in subscription EdTech
Publication . Gavrilova, Liubov; António, Nuno; NOVA Information Management School (NOVA IMS); Information Management Research Center (MagIC) - NOVA Information Management School; Springer Boston
Subscription-based educational technology (EdTech) companies experience significant revenue loss from student churn, making retention vital. Retaining engaged learners is generally more cost-effective than acquiring new ones. This study develops a predictive churn model using anonymised student data from an EdTech platform for children’s English learning. Following a CRISP-DM process, student activity, engagement, and financial features were engineered across multiple time windows to capture evolving student behaviours. Four machine learning algorithms-logistic regression, random forest, neural networks, and XGBoost (XGB)—were trained and compared. The optimised XGB model achieved the best performance, with approximately 0.84 accuracy, 0.63 F1 score, 0.85 recall, and the area under the curve (AUC) of 0.85 on test data, effectively identifying likely churners. Shapley Additive exPlanations (SHAP) based analysis revealed that engagement metrics, particularly the number of paid classes (both current and mean over 8 weeks), total learning time, engagement at gamified features, and student tenure, were the most influential predictors, confirming that highly engaged students are less likely to churn. This interpretable model provides actionable insights for retention strategies by predicting individual churn risk and highlighting key engagement drivers. In practice, even a 1% monthly reduction in churn could translate into multi-million-dollar annual savings for subscription EdTech providers. Overall, this research extends churn prediction into the EdTech domain, demonstrates the value of long-term engagement features, and applies explainable AI to enhance model transparency, thereby supporting its practical adoption.