http://www.morelab.deusto.es/publications/1/MORElab publicationsen-ENMon, 09 Mar 2026 02:45:27 +0000http://www.morelab.deusto.es/publications/info/towards-a-framework-for-intelligent-sampling-comprehensive-review-of-challenges-ai-techniques-and-tools/http://www.morelab.deusto.es/publications/info/towards-a-framework-for-intelligent-sampling-comprehensive-review-of-challenges-ai-techniques-and-tools/http://www.morelab.deusto.es/publications/info/2025-ieee-international-conference-on-smart-computing-smartcomp-2025-/http://www.morelab.deusto.es/publications/info/2025-ieee-international-conference-on-smart-computing-smartcomp-2025-/http://www.morelab.deusto.es/publications/info/greendai-towards-an-observability-tool-for-sustainable-green-distributed-artificial-intelligence/The growth of intelligent systems based on distributed artificial intelligence has generated an urgent need to assess their environmental impact and sustainability. The expansion of AI technologies has driven an increase in demands for computing resources and has augmented the need to adapt to more efficient forms of computing. such as distributed AI or Edge AI. Additionally, there is growing concern about energy consumption and the environmental footprint of these technologies. In this paper, our main contribution is the definition of a Distributed Artificial Intelligence systems observability tool architecture for green and sustainability perspectives. This tool is designed to report adequate KPIs by measuring the energy consumption, data usage, computational efficiency and carbon footprint of decentralized AI based components. This innovative approach not only promotes the development of more sustainable technologies but also encourages transparency and responsibility in the sustainable use of distributed and large-scale AI systems. To validate the approach, its feasibility and integration have been analyzed in an experimental use case of a Distributed AI system. The use case is a Federated Machine Learning based system, in which the benefits of reporting energy efficiency and sustainability metrics are analyzed.http://www.morelab.deusto.es/publications/info/greendai-towards-an-observability-tool-for-sustainable-green-distributed-artificial-intelligence/http://www.morelab.deusto.es/publications/info/2025-9th-international-symposium-on-innovative-approaches-in-smart-technologies-isas-2025-/http://www.morelab.deusto.es/publications/info/2025-9th-international-symposium-on-innovative-approaches-in-smart-technologies-isas-2025-/http://www.morelab.deusto.es/publications/info/greencrowd-platform-performance-testing-toolkit/This repository provides a reproducible deployment and load testing toolkit for the GREENCROWD platform — an open-source citizen science and gamification system. It includes Docker-based infrastructure (docker-compose.yml), a pre-configured .env template, deployment instructions, and a Locust script for benchmarking API performance. The toolkit enables researchers and developers to quickly deploy GREENCROWD and simulate realistic user interactions under scalable conditions, supporting evaluation of system robustness and scalability.http://www.morelab.deusto.es/publications/info/greencrowd-platform-performance-testing-toolkit/http://www.morelab.deusto.es/publications/info/-2025--/http://www.morelab.deusto.es/publications/info/-2025--/http://www.morelab.deusto.es/publications/info/detecting-suicidal-ideation-on-social-media-using-large-language-models-with-zero-shot-prompting/http://www.morelab.deusto.es/publications/info/detecting-suicidal-ideation-on-social-media-using-large-language-models-with-zero-shot-prompting/http://www.morelab.deusto.es/publications/info/categorizing-and-assessing-aspects-of-suicidal-ideation-detection-approaches-a-systematic-review/Suicide remains a critical global issue and one of the leading causes of death worldwide. As this problem grows, the need for effective prevention strategies becomes increasingly urgent. Social networks and online platforms, such as Twitter, have emerged as spaces where people openly share their thoughts and emotions, including negative feelings, reflections on life, and even suicidal thoughts. This makes social media data an important resource for efforts to detect and reduce the risk of suicide. This systematic review examines 92 studies published between 2018 and 2024 on the detection of suicidal ideation. The studies are categorized using a multidimensional framework that considers three key aspects: the platforms used for data collection, the analytical techniques applied, and the specific features employed to identify suicidal ideation. By exploring these dimensions, the review highlights existing gaps and limitations in current methods, offering insights to guide future research and improve strategies for suicide prevention.http://www.morelab.deusto.es/publications/info/categorizing-and-assessing-aspects-of-suicidal-ideation-detection-approaches-a-systematic-review/http://www.morelab.deusto.es/publications/info/computers-in-human-behavior-reports-2025-19-/http://www.morelab.deusto.es/publications/info/computers-in-human-behavior-reports-2025-19-/http://www.morelab.deusto.es/publications/info/predictive-assessment-of-eating-disorder-risk-and-recovery-uncovering-the-effectiveness-of-questionnaires-and-influencing-characteristics/http://www.morelab.deusto.es/publications/info/predictive-assessment-of-eating-disorder-risk-and-recovery-uncovering-the-effectiveness-of-questionnaires-and-influencing-characteristics/http://www.morelab.deusto.es/publications/info/computational-and-structural-biotechnology-journal-2025-28-/http://www.morelab.deusto.es/publications/info/computational-and-structural-biotechnology-journal-2025-28-/http://www.morelab.deusto.es/publications/info/democratizing-co-production-of-thematic-co-explorations-for-citizen-observatories/Citizen Observatories are a promising instrument to drive societal behaviour change towards greener more sustainable practices. However, assembling Citizen Observatories is not easy, since apart from the continuous engagement of their participants, there is the need to have some specialized domain and technical knowledge. Data quality, continuous engagement and retention and factual impact into decision making are three usual roadblocks which impend a wider adoption of this practice. This paper explains how GREENGAGE project aims to democratize the co-production of thematic co-explorations and overcome those barriers.http://www.morelab.deusto.es/publications/info/democratizing-co-production-of-thematic-co-explorations-for-citizen-observatories/http://www.morelab.deusto.es/publications/info/analysing-the-impact-of-images-and-text-for-predicting-human-creativity-through-encoders/http://www.morelab.deusto.es/publications/info/analysing-the-impact-of-images-and-text-for-predicting-human-creativity-through-encoders/http://www.morelab.deusto.es/publications/info/proceedings-of-the-11th-international-conference-on-information-and-communication-technologies-for-ageing-well-and-e-health-2025-/http://www.morelab.deusto.es/publications/info/proceedings-of-the-11th-international-conference-on-information-and-communication-technologies-for-ageing-well-and-e-health-2025-/http://www.morelab.deusto.es/publications/info/tailoring-digital-middleware-for-rural-needs-insights-from-user-feedback-analysis/The digitalization of rural areas is essential for bridging the socio-economic gap between urban and rural communities. However, the adoption of digital technologies in rural environments faces significant barriers, including limited infrastructure, technical knowledge requirements, and user accessibility challenges. For this reason, middleware platforms aim to enhance digital infrastructure and interoperability in rural areas by providing user-friendly digital services. This paper presents an evaluation of the feedback obtained on the AURORAL middleware for rural areas to understand its effectiveness and identify areas for improvement. By analyzing insights from focus groups and questionnaires with eight pilots, we aim to highlight the real-world challenges and benefits experienced by participants when interacting with the middleware. This feedback is crucial for refining such solutions to meet better the needs of diverse stakeholders, including those providing and accessing services, and ensuring its successful adoption. Moreover, the findings of this study emphasize the importance of user-centered design in overcoming digital barriers, enhancing usability and accessibility, and promoting sustainable development in rural areas, ultimately contributing to a more inclusive digital transformation.http://www.morelab.deusto.es/publications/info/tailoring-digital-middleware-for-rural-needs-insights-from-user-feedback-analysis/http://www.morelab.deusto.es/publications/info/proceedings-of-the-international-conference-on-ubiquitous-computing-and-ambient-intelligence-ucami-2024-2024-1212-/http://www.morelab.deusto.es/publications/info/proceedings-of-the-international-conference-on-ubiquitous-computing-and-ambient-intelligence-ucami-2024-2024-1212-/http://www.morelab.deusto.es/publications/info/machine-learning-approaches-for-predicting-heart-failure-readmissions/Abstract Purpose This study aims to develop and evaluate machine learning (ML) models to predict the likelihood of hospital readmission within 30 days after discharge for patients with heart failure (HF). The goal is to compare the predictive accuracy of ML models with traditional methods such as those based on Cox proportional hazards and logistic regression, to improve clinical outcomes and reduce hospital costs. Methods We conducted a prospective cohort study of patients discharged from five hospitals following admission for HF. Data were collected on variables including sociodemographic characteristics, medical history, admission details, patient-reported outcomes, and clinical parameters. ML techniques were employed to analyse the data and predict readmission risk, incorporating strategies to handle class imbalance and missing data. Model performance was assessed based on accuracy, sensitivity, specificity, area under the receiver operating characteristic curve (AUC), and F1 score. Results Ensemble methods with Synthetic Minority Over-sampling Technique balancing and bagging improved the predictive performance of ML models compared with traditional models. The best-performing ensemble model, using decision trees, Gaussian Naïve Bayes, and neural networks, achieved an AUC of 0.81. In contrast, Cox and logistic regression models showed significantly poorer performance (AUC of 0.58 and 0.50, respectively). SHapley Additive exPlanations analysis revealed that frailty, anxiety, and depression were critical in predicting readmission. Conclusion ML models, particularly those using ensemble methods, significantly outperform traditional models in predicting short-term readmission for patients with HF. These findings highlight the potential of ML to improve clinical decision-making and resource allocation in HF management.http://www.morelab.deusto.es/publications/info/machine-learning-approaches-for-predicting-heart-failure-readmissions/http://www.morelab.deusto.es/publications/info/postgraduate-medical-journal-2025--/http://www.morelab.deusto.es/publications/info/postgraduate-medical-journal-2025--/http://www.morelab.deusto.es/publications/info/development-of-an-open-source-iot-blockchain-platform-for-traceability-of-fresh-products-from-farm-to-fork/http://www.morelab.deusto.es/publications/info/development-of-an-open-source-iot-blockchain-platform-for-traceability-of-fresh-products-from-farm-to-fork/http://www.morelab.deusto.es/publications/info/blockchain-and-applications-5th-international-congress-2023-778-/http://www.morelab.deusto.es/publications/info/blockchain-and-applications-5th-international-congress-2023-778-/http://www.morelab.deusto.es/publications/info/smart-home-assisted-anomaly-detection-system-for-older-adults-a-deep-learning-approach-with-a-comprehensive-set-of-daily-activities/Abstract Smart homes have the potential to enable remote monitoring of the health and well-being of older adults, leading to improved health outcomes and increased independence. However, current approaches only consider a limited set of daily activities and do not combine data from individuals. In this work, we propose the use of deep learning techniques to model behavior at the population level and detect significant deviations (i.e., anomalies) while taking into account the whole set of daily activities (41). We detect and visualize daily routine patterns, train a set of recurrent neural networks for behavior modelling with next-day prediction, and model errors with a normal distribution to identify significant deviations while considering the temporal component. Clustering of daily routines achieves a silhouette score of 0.18 and the best model obtains a mean squared error in next day routine prediction of 4.38%. The mean number of deviated activities for the anomalies in the train and test set are 3.6 and 3.0, respectively, with more than 60% of anomalies involving three or more deviated activities in the test set. The methodology is scalable and can incorporate additional activities into the analysis. Graphical abstract A comprehensive activity monitoring and anomaly detection system for older adults, using sensor data, predictive modeling, and statistical analysis to alert health professionals of irregular behaviors.http://www.morelab.deusto.es/publications/info/smart-home-assisted-anomaly-detection-system-for-older-adults-a-deep-learning-approach-with-a-comprehensive-set-of-daily-activities/http://www.morelab.deusto.es/publications/info/medical-biological-engineering-computing-2025--/http://www.morelab.deusto.es/publications/info/medical-biological-engineering-computing-2025--/http://www.morelab.deusto.es/publications/info/an-image-based-sensor-system-for-low-cost-airborne-particle-detection-in-citizen-science-air-quality-monitoring/Air pollution poses significant public health risks, necessitating accurate and efficient monitoring of particulate matter (PM). These organic compounds may be released from natural sources like trees and vegetation, as well as from anthropogenic, or human-made sources including industrial activities and motor vehicle emissions. Therefore, measuring PM concentrations is paramount to understanding people's exposure levels to pollutants. This paper introduces a novel image processing technique utilizing photographs/pictures of Do-it-Yourself (DiY) sensors for the detection and quantification of PM10 particles, enhancing community involvement and data collection accuracy in Citizen Science (CS) projects. A synthetic data generation algorithm was developed to overcome the challenge of data scarcity commonly associated with citizen-based data collection to validate the image processing technique. This algorithm generates images by precisely defining parameters such as image resolution, image dimension, and PM airborne particle density. To ensure these synthetic images mimic real-world conditions, variations like Gaussian noise, focus blur, and white balance adjustments and combinations were introduced, simulating the environmental and technical factors affecting image quality in typical smartphone digital cameras. The detection algorithm for PM10 particles demonstrates robust performance across varying levels of noise, maintaining effectiveness in realistic mobile imaging conditions. Therefore, the methodology retains sufficient accuracy, suggesting its practical applicability for environmental monitoring in diverse real-world conditions using mobile devices.http://www.morelab.deusto.es/publications/info/an-image-based-sensor-system-for-low-cost-airborne-particle-detection-in-citizen-science-air-quality-monitoring/http://www.morelab.deusto.es/publications/info/sensors-basel-switzerland-2024-24-19/http://www.morelab.deusto.es/publications/info/sensors-basel-switzerland-2024-24-19/http://www.morelab.deusto.es/publications/info/the-role-of-iot-devices-in-sustainable-car-expenses-in-the-context-of-the-intelligent-mobility-a-comparative-approach/Connected cars have often been defined as vehicles that can provide some services and information without human intervention. Several scholars have examined the factors that promote the purchase or adoption of such augmented vehicles. However, little emphasis has been placed on the determinants for reducing car expenditures when a driver owns a car with an Internet of Things (IoT) device or a smart assistant in the context of smart mobility. Therefore, this article analyzes whether emerging technology such as IoT plays a key factor for a driver concerning the expenses related to the car (e.g., insurance, maintenance, and repairs). To this extent, a methodology based on exploratory (i) and confirmatory analysis (ii) was carried out. The authors initially conducted an exploratory phase by means of a Delphi method in which a group of vehicle experts (n = 25) were recruited to give their opinions and reach an agreement defining the determinants that they believed affected vehicle expenditures the most. Secondly, and taking into consideration that the salient determinant from the Delphi method was the use of technology and the warnings and alerts it triggers, a questionnaire was delivered to 556 drivers to analyze the everyday spending on their cars. Specifically, the survey aimed to compare the responses of people who own connected cars or have any kind of built-in IoT infrastructure (n = 302) with those of people with non-connected cars (n = 254). The main conclusion obtained for this latter approach was that drivers with a connected car have remarkably lower car expenses than those driving a conventional car.http://www.morelab.deusto.es/publications/info/the-role-of-iot-devices-in-sustainable-car-expenses-in-the-context-of-the-intelligent-mobility-a-comparative-approach/http://www.morelab.deusto.es/publications/info/applied-sciences-2022-12-3/http://www.morelab.deusto.es/publications/info/applied-sciences-2022-12-3/http://www.morelab.deusto.es/publications/info/human-in-the-loop-machine-learning-reconceptualizing-the-role-of-the-user-in-interactive-approaches/The rise of intelligent systems and smart spaces has opened up new opportunities for human–machine collaborations. Interactive Machine Learning (IML) contribute to fostering such collaborations. Nonetheless, IML solutions tend to overlook critical factors such as the timing, frequency and workload that drive this interaction and are vital to adapting these systems to users’ goals and engagement. To address this gap, this work explores users’ expectations towards IML solutions in the context of an interactive hydration monitoring system for the workplace, which represents a challenging environment to implement intelligent solutions that can collaborate with individuals. The proposed system involves users in the learning process by providing feedback on the success of detecting their drinking gestures and enabling them to contribute with additional examples of their data. A qualitative study was conducted to evaluate this use case, where participants completed specific tasks with varying levels of involvement. This study provides promising insights into the potential of placing the Human-in-the-Loop (HitL) to adapt and reconceptualize the users’ role in interactive solutions, highlighting the importance of considering human factors in designing more effective and flexible collaborative systems between humans and machines.http://www.morelab.deusto.es/publications/info/human-in-the-loop-machine-learning-reconceptualizing-the-role-of-the-user-in-interactive-approaches/http://www.morelab.deusto.es/publications/info/understanding-the-factors-that-affect-households-investment-decisions-required-by-the-energy-transition/In energy systems’ economic models, people’s behaviour is often underestimated, and they are generally unaware of how habits impact energy efficiency. Improving efficiency is challenging, and recommendations alone may not be sufficient. Changing behaviour requires understanding the direct impact of needs and habits on energy efficiency. This research introduces a methodology that retrieves human expert knowledge from four key aspects of the current energy transition: everyday appliances, buildings, mobility, flexibility, and energy efficiency. The aim is to examine the causal relationship between energy consumption and human behaviour, gaining a deeper understanding of the links among the factors that drive final energy consumers to change habits through the adoption of energy-saving measures. Working in collaboration with expert panels, this study provides a methodology for extracting expert human knowledge based on a set of future energy transition scenarios aligned with the achievement of the Paris Agreement, a taxonomy of 32 factors that have a strong influence on households’ investment decisions, and the results of a survey that characterises the European population through the 32-factor taxonomy and some socioeconomic conditions. In addition, the survey included a sample of the Latin American population to analyse how socioeconomic conditions (region, education, gender, etc.) influence the prioritisation of these factors. We discuss the high priority given to competence and autonomy over financial factors by inhabitants of the European Union residential sector. We provide an analysis of the factors through which other similar projects are focused and on which we converge. In addition, we contribute by presenting the hierarchy of priorities assigned by people. This highlights the importance for policymakers to take these aspects seriously when implementing energy policy interventions that go beyond purely financial measures and fiscal incentives.http://www.morelab.deusto.es/publications/info/understanding-the-factors-that-affect-households-investment-decisions-required-by-the-energy-transition/http://www.morelab.deusto.es/publications/info/plos-one-2024-19-3/http://www.morelab.deusto.es/publications/info/plos-one-2024-19-3/http://www.morelab.deusto.es/publications/info/internet-of-things-2024-25-/http://www.morelab.deusto.es/publications/info/internet-of-things-2024-25-/