IRIS Università degli Studi dell'Aquila
Not a member yet
68355 research outputs found
Sort by
Investigating the effects of hyperparameter sensitivity on machine learning algorithms for PV forecasting
No full textMachine Learning (ML) models have been introduced in the past, and users have debated whether to tune the hyperparameters of the models. This study investigates the effects of tuning the hyperparameters of the ML models and summarizes the models that are most sensitive to hyperparameter tuning. This study leveraged the historic energy production
data of two already operational PV plants. Four state-of-the-art ML models, namely Decision Trees (DT), Random Forest (RF), K-Nearest Neighbors (KNN), and Support Vector Regression (SVR) were investigated. All the ML models were trained with the same training features (meteorological estimates) obtained from the National Aeronautics and Space
Administration’s (NASA) Power project, with the daily PV energy production selected as the target variable. Models were developed and executed with default and tuned hyperparameters using an 85-15% traintest split. The results revealed that all the models showed improved performance with the tuned hyperparameters. However, the DT and SVR
models depicted significantly improved RMSE after tuning of the hyperparameters. The RMSE of DT improved from 111 kWh/d to 75 kWh/d for one plant and from 442 kWh/d to 270 kWh/d for the second plant after tuning the hyperparameters. Similarly, the RMSE of SVR improved from 59 kWh/d to 50 kWh/d in the first case, and in the second case, the improvement of RMSE from 536 kWh/d to 294 kWh/d was observed. The efficiency of the RF and KNN models also improved to some extent after tuning, but the RMSE closely agreed with the default hyperparameters in one case study, making the RF and KNN less prone to hyperparameter sensitivity. This study concluded with the finding that it is
necessary to tune the hyperparameters of the DT and SVR models, specifically for energy forecasting. Moreover, the results of this study also highlight the significance of meteorological estimates from NASA’s Power project, as models successfully discerned the complex energy forecast patterns. The dataset is deemed suitable for energy forecasting for areas with sparse ground-based observatories and may serve as a baseline dataset
for training the ML models
Ethical, Legal, and Societal Dimensions of AI-Driven Social Robots in Elderly Healthcare
No full textAI-driven social robots in elderly healthcare are one of the most important applications to increase the quality of life by promoting activities of daily living, exercises to improve memory and cognitive functioning. On the other hand, these new levels of innovation also raise troubling questions, including those related to ethics, legal compliance, and societal issues with a particular focus around privacy and data security. This paper discusses these dimensions from the point of view of human-robot interaction (HRI) augmented by AI. We review and analyze the literature, regulations, and standards to identify gaps where pragmatic improvements can be instituted, offering guidelines for a responsible ethical implementation of AI-driven social robots assisting the elderly. Additionally, the ethical implications of anthropomorphism, where robots are endowed with human-like traits, are examined, particularly in terms of their emotional impact on users. Our aim is to provide a comprehensive understanding of the current landscape and identify areas for improvement, ensuring that the deployment of social robots in elderly care respects ethical standards and complies with regulatory requirements. By addressing these challenges, we can fulfil the positive potential of social robots in improving the quality of life for the elderly.AI-driven social robots in elderly healthcare are one of the most important applications to increase the quality of life by promoting activities of daily living, exercises to improve memory and cognitive functioning. On the other hand, these new levels of innovation also raise troubling questions, including those related to ethics, legal compliance, and societal issues with a particular focus around privacy and data security. This paper discusses these dimensions from the point of view of human-robot interaction (HRI) augmented by AI. We review and analyze the literature, regulations, and standards to identify gaps where pragmatic improvements can be instituted, offering guidelines for a responsible ethical implementation of AI-driven social robots assisting the elderly. Additionally, the ethical implications of anthropomorphism, where robots are endowed with human-like traits, are examined, particularly in terms of their emotional impact on users. Our aim is to provide a comprehensive understanding of the current landscape and identify areas for improvement, ensuring that the deployment of social robots in elderly care respects ethical standards and complies with regulatory requirements. By addressing these challenges, we can fulfil the positive potential of social robots in improving the quality of life for the elderly
Recycling of end-of-life solar panels: Focusing on the pyrolysis conversion of back sheet from a micro perspective
No full textNarcissism and the risk of exercise addiction in youth: the impact of problematic social media use and fitspiration exposure
No full textExercise represents precious tool in multimodal interventions for countering pathological conditions and promoting individual well-being. However, it might evolve into an addiction when people overinvest and prioritize training, neglecting other areas of their lives. This study investigates the association between both grandiose and vulnerable narcissism and the risk of exercise addiction (EA), addressing the involvement of problematic social media use (PSMU) and fitspiration exposure (i.e. the exposure to a specific kind of social media content aimed to motivate towards fitness and healthy lifestyle ostentatiously). The study employs an online cross-sectional design with 173 emerging adults (Mage = 21.96 years; SDage = 2.37 years; rangeage 18–25 years). The mediation analysis indicates that both grandiose and vulnerable narcissism indirectly affect the risk of EA through the sequential mediating effect of PSMU and fitspiration exposure. These findings highlight the role of the digital environment and socio-cultural pressures on the risk of EA. Practical implications include the need for targeted social media literacy campaigns to promote healthier engagement with online platforms and foster exercise practices that prioritize mental and physical well-being over appearance-driven goals. Limitations and future research directions are discussed
Knowledge, Barriers, and Future Directions of Vestibular Rehabilitation Practice in Neurorehabilitation: An Italian Survey
Get PDFBackground/Objectives: Vestibular rehabilitation, an evidence-based physical
therapy approach, plays a crucial role in managing and recovering from gaze and balance
disorders, including those of central origin. This study, targeted at the community of Italian
healthcare practitioners, is vital in understanding the application of vestibular rehabilitation
in neurological disorders and in identifying knowledge gaps, barriers, and future directions. Methods: This is a cross-sectional study directed at healthcare professionals involved
in neurorehabilitation in Italy. The survey consisted of 29 items grouped in 4 sections,
which was estimated to take approximately 10 min to complete. The questions covered
socio-demographic information, professional information, clinical practice, and future
perspectives on vestibular rehabilitation. Results: Out of the 435 respondents, 290 completed the survey. Most of the respondents reported either no (32.87%) or little (42.91%)
experience in vestibular rehabilitation. However, most participants (72.98%) recognized
the importance of vestibular rehabilitation in treating neurological disorders. The most
common condition treated was stroke (46.39%), while balance training (52.69%) and visual
input exercises (26.35%) were the two most frequently used strategies. The main barriers
to implementing vestibular rehabilitation in clinical practice were equipment cost and
insufficient skills. Conclusions: Vestibular physical therapy is a promising complementary
approach in neurorehabilitation. However, the study reveals a perceived lack of basic
training in vestibular assessment and therapy. This suggests that more efforts are needed
to bridge this knowledge gap and make necessary equipment more accessible
Predictors of Nurses' Job Satisfaction in Home Care Settings: Findings From the AIDOMUS-IT Study
No full textIntroduction: Nurses' job satisfaction in hospitals is fundamental for the quality of care and the safety of patients. However, sociodemographic trends require moving care to patients' homes, and the predictors of job satisfaction for nurses working in the home care settings remain largely unknown. Therefore, the aim of this study was to investigate job satisfaction of nurses working in Italian home care settings and its determinants. Design: Multicenter observational cross-sectional study. Methods: This study was conducted in the districts of 70 local health authorities in Italy. Data on the characteristics of the organization and nurses were collected. Nursing job satisfaction was evaluated on a four-point scale ranging from "very satisfied" to "very dissatisfied." Additionally, the following variables were assessed: workload, quality of leadership, work-private life conflict, burnout symptoms, possibility for development, staffing and resource adequacy, nurse manager ability, safety climate, and teamwork climate. A logistic regression analysis was conducted to identify factors influencing job satisfaction. Results: Only organizational variables had a predictive value for nurses' job satisfaction. Workload (OR = 1.01; p = 0.033), work-private life conflict (OR = 1.02; p < 0.001), burnout (OR = 1.02; p < 0.001), and staffing inadequacy (OR = 1.44; p = 0.003) predicted higher levels of nurse dissatisfaction. Instead, high-quality leadership (OR = 0.981; p < 0.001), possibility for development (OR = 0.973; p < 0.001), and good teamwork climate (OR = 0.994; p = 0.003) were predictors of better levels of satisfaction. Conclusions: This study suggested that home care nurses are generally satisfied with their jobs. To enhance job satisfaction, it is essential to improve nurses' work environment, the leadership quality and ensure professional development. Clinical relevance: Our results are globally relevant as they contribute to the limited evidence available on this topic in home care settings. This study emphasizes the need of measuring nurses' job satisfaction and implementing interventions to promote healthy work environments
Towards the Digital Twin of a Deep Underground Laboratory. The Information Modelling of the LNGS Experiment Facilities
No full textDigital Twin technology within the AEC industry is a crucial topic for the effective management of the built environment. The PNRR-funded research project Tecnodigit also focuses on this strategy, proposing a more effective integration of sensor data into BIM models to guarantee a higher level of building safety and sustainability. Among the case studies involved in the research project, the underground facilities of National Gran Sasso Laboratories represent an exceptional example not only for their unusual location, but also for their unconventional building typology mixing buildings and infrastructural features. The challenge of developing a federated BIM model of the central tunnel containing the experiments within for physics particles involves the overcoming of several criticalities
that are described and highlighted in the paper, along with the proposal of possible solutions that represent the first phases for the application of a DT system
Timing and Safety of Anticoagulation Reinitiation After Intracranial Hemorrhage in Patients With Mechanical Valves
Get PDFBACKGROUND AND OBJECTIVES: In patients with mechanical heart valves (MHVs), anticoagulation (AC) interruption after intracranial hemorrhage (ICH) poses a clinical dilemma because of competing risks of ischemic complications and hemorrhagic recurrence. To date, the optimal timing for resuming vitamin K antagonists (VKAs) remains unclear. The aim of this meta-analysis was to quantify the risks of ischemic stroke and recurrent ICH associated with VKA resumption in this population and explore the temporal risk dynamics. METHODS: We systematically searched PubMed, Embase, and Cochrane Library from inception to December 2023 for studies reporting ischemic or hemorrhagic outcomes in adults with MHVs who experienced ICH and were considered for VKA resumption. Primary outcomes were ischemic stroke before AC resumption and recurrent ICH after AC resumption. Random-effects meta-analyses were performed. Meta-regressions assessed whether timing of resumption influenced risk. Risk trajectories were estimated using a model-based approach. RESULTS: Nine studies were included, comprising 435 patients with MHVs with confirmed ICH included in the pooled analysis. The mean age ranged from 54.1 to 75 years; 31.3% were female. The pooled incidence of recurrent ICH after AC reinitiation was 11.4% (95% CI 8.2-15.6; I2 = 0%), the incidence of ischemic stroke during AC suspension was 6.1% (95% CI 4.1-8.9; I2 = 0%), valve thrombosis occurred in 3.3% (95% CI 1.9-5.6; I2 = 0%), and mortality occurred in 4.9% (95% CI 2.0-11.5; I2 = 37%). Meta-regression demonstrated a significant inverse association between time to AC resumption and risk of recurrent ICH (regression coefficient -0.039; 95% CI -0.093 to 0.015; p = 0.13), corresponding to an approximate 50% relative reduction in risk at 11 days after ICH. No significant time-dependent association was observed for ischemic stroke (coefficient -0.013; 95% CI -0.065 to 0.039; p = 0.61). DISCUSSION: In patients with MHVs who experienced an ICH, this meta-analysis found that resumption of AC was associated with a recurrent ICH rate of 11.4% and an ischemic stroke rate of 6.1% during AC suspension. Meta-regression suggested a lower risk of recurrent ICH with later AC resumption, with a potential risk reduction at approximately 11 days after ICH. No time-dependent increase in ischemic stroke was observed. Limitations include the retrospective design of most studies and heterogeneous AC timing across cohorts
