- Research Article
- 10.1016/j.cose.2026.104848
Assessing the attack surface of space organizations: A data-driven analysis
- May 01, 2026
- Computers & Security
- Francesco Casaril + 1 more +1
Publications from 2021 to 2026
Showing 10 of 531 papers
Assessing the attack surface of space organizations: A data-driven analysis
HPVoice: a single-blind three-arm RCT on social media communication strategies among adolescents in Tuscany, Italy.
Cervical cancer is the first cancer potentially eliminable through widespread HPV vaccination. In Italy, despite free access, vaccine coverage among adolescents remains low. Psychological factors such as information avoidance (IA), the tendency to ignore threatening information, can influence vaccine decisions and contribute to vaccine hesitancy (VH). While social media offers new avenues for health communication, their effectiveness in reducing VH is uncertain. This study investigates the impact of Instagram-based communication strategies on adolescents' HPV knowledge and VH. A three-arm randomized controlled trial was conducted in 22 high school classes in Tuscany, involving 526 students. Classes were randomized to follow one of three Instagram pages for one month: a neutral control, an information-based account, or a bias-aware account designed to address cognitive biases related to vaccination decisions. Data were collected through online surveys at three time points: baseline (T1), after the intervention (T2), followed by in-class lesson, and at three-month follow-up (T3). HPV knowledge and VH were assessed at all time points, whereas cognitive biases, including IA, were measured only at T1. The final analysis included 301 students (mean age 15.2, 74% female). Overall, no significant changes in VH or knowledge were found between T1 and T2 overall. However, in students with low IA the info-based page reduced VH significantly compared to the control (-6.9%, p=.009). In contrast, among high-IA students, the same intervention increased VH (estimated: +8.9%, p=.002). Willingness to seek information increased in the info group among low-IA students (logg odds β=+1.639 vs. control, p=.034). At T3, we measured significant increase in knowledge compared to T1 despite the smaller sample collected. Social media campaigns showed limited effects on VH reduction in adolescents and can even backfire among high-IA individuals. Future strategies should incorporate personalized, bias-aware and interactive tools to address VH and HPV vaccine uptake effectively.
Read moreAccent Discrimination in Entrepreneurial Fundraising
Accent is a powerful, yet still socially acceptable, basis of discrimination that shapes interpersonal evaluations and access to opportunities, potentially disadvantaging entrepreneurs seeking funding. Bridging role congruity theory with sociolinguistic literature, we hypothesize that entrepreneurs whose regional accents align with the stereotypical entrepreneur are more likely to secure funding. We test this hypothesis in entrepreneurial pitching contexts using an observational study and two pre-registered experiments that manipulate entrepreneurs’ accents in pitch videos through a professional voice actor and AI. Across the UK, Italy, and the United States, the results support our hypothesis, providing robust evidence of accent-based discrimination in entrepreneurial fundraising.
Read moreIntegrating isotope hydrology and hydrogeochemistry to assess recharge processes and vulnerability of drinking water springs: the case of Capannori (Tuscany, central Italy)
Ongoing climate change and increasing anthropogenic pressures are intensifying the challenges associated with the sustainable management of groundwater resources, particularly in Mediterranean regions characterized by pronounced climatic seasonality and recognized as climate change hot spots. Groundwater springs often represent a critical resource for local communities, serving both as a source of high-quality drinking water and as a key component of socio-ecological systems. Understanding the functioning, vulnerability, and resilience of the groundwater systems feeding these springs is therefore essential for effective and sustainable water management.This study presents the first results of an ongoing investigation in the Municipality of Capannori (Lucca, Tuscany, central Italy), where publicly accessible springs are widely used for domestic water supply and have become the focus of increasing conservation efforts by local authorities. The collection of water from local springs represents a shared practice in this region that fosters a strong sense of community and reinforces the concept of water as a common good. Ensuring the sustainable management of these resources, therefore, requires a comprehensive understanding of groundwater systems, achievable through multidisciplinary approaches integrating isotopic fingerprinting with geochemical and hydrogeological tools.A two-year monitoring programme started in April 2024 and involves 19 groundwater springs distributed across three hydrogeological sectors. Groundwater samples were collected for chemical and isotopic analysis (δ18O and δ2H), while temperature, pH, electrical conductivity and redox potential were measured in the field. Tritium and δ34S-δ18O-SO4 were determined on selected samples, and compositional data analysis (CoDA) was applied to the chemical dataset. In addition, a rain sampler was installed in February 2024 to collect monthly precipitation samples for isotopic analysis.The results highlight pronounced chemical and isotopic heterogeneity among springs across the three hydrogeological sectors, reflecting the complexity of the groundwater systems involved. This heterogeneity points to distinct flow paths, water-rock interaction processes, and recharge dynamics, and suggests potentially different sensitivities of individual springs to hydroclimatic variability and anthropogenic pressures. From a management perspective, these differences imply that springs commonly perceived as part of a single resource may in fact exhibit contrasting levels of vulnerability under ongoing and future environmental change.Stable water isotope data, together with deuterium excess, provide robust constraints on the recharge elevations of the aquifers feeding the springs, allowing the identification of recharge areas and the possible extent of recharge catchments. At the same time, part of the observed isotopic variability may reflect a climatic signal related to recharge seasonality rather than elevation alone. However, seasonal isotopic shifts were negligible across all springs, indicating well-mixed recharge systems and relatively slow groundwater circulation that dampens the pronounced isotopic variability of precipitation. Consistently, tritium values show no significant differences among springs (2.5-2.9 TU), indicating young groundwater with residence times not exceeding 5-10 years.Overall, this study demonstrates how integrated isotopic and geochemical approaches can provide process-based insights that are directly relevant for groundwater protection and management under changing hydroclimatic conditions.
Read moreUnderstanding Energy Efficiency of AI Deployments in IoT-Driven Smart Cities
The pervasive adoption of AI and AIoT applications at the network edge presents both opportunities and challenges for smart cities. With a focus on the energy efficiency of AI in urban environments, this paper provides a systematic comparative analysis of representative edge hardware platforms, i.e., embedded GPUs, FPGAs, and ultra-low-power microcontroller-/sensor-class devices, assessing their suitability for AI workloads in IoT-driven smart city infrastructures. The evaluation, based on direct characterization of diverse neural networks and relevant datasets, quantifies computational performance and energy behavior through inference latency, throughput, and energy/per inference measurements. Across the evaluated network–board pairs, the measured inference power spans several orders of magnitude, ranging from 0.1–10 mW for ultra-low-power Intelligent Sensor Processing Units (ISPUs) up to 1–10 W for embedded GPUs, highlighting the wide design space between the least and most power-demanding configurations. Results indicate that embedded GPUs provide a favorable performance-to-power ratio for computationally intensive workloads, while MCU/ISPU-class solutions, despite throughput limitations, offer compelling advantages in ultra-low-power scenarios when combined with quantization and pruning, making them well-suited for distributed sensing and actuation typical of smart city deployments. Overall, this comparative analysis guides hardware selection for heterogeneous, sustainable AI-enabled urban services.
Read moreIntegrating neural, physiological, and interoceptive measures in social interaction
Social interaction is not simply the meeting of two brains, but the emergent product of continuously coupled brain-body systems within and between individuals. Hyper scanning studies have shown that interacting partners exhibit inter-brain synchrony during cooperation, conversation, and joint attention, consolidating neural coupling as a marker of shared engagement. Yet neural data alone cannot tell us whether individuals are aligned in their emotions, arousal, or regulatory strategies, nor how these states evolve in real time. In parallel, work on physiological synchrony has revealed that partners' autonomic signals also converge during interaction, tracking rapport, cohesion, and cooperative success. Interoception, the sensing and representation of internal bodily signals, provides a conceptual bridge between these levels. Atypical interoceptive processing is increasingly implicated across conditions marked by social difficulties, and emerging hyper scanning studies show that directing attention to bodily rhythms can enhance both autonomic and neural alignment. In this Perspective, I argue for a framework in which social behavior arises from multilevel coupling among neural activity, autonomic physiology, and interoceptive processes. I outline how multimodal hyperscanning in naturalistic tasks, combined with temporally sensitive analyses, can clarify when and how these levels interact, from dyads to groups. Finally, I discuss clinical and developmental horizons in which multimodal synchrony serves both as a mechanistic readout and a potential target for interventions that stabilize internal dynamics to support social connection.
Read moreCognitive Modelism
Abstract Structures are ubiquitous in mathematics. But how should they be understood? Modelists claim they are model-theoretic structures. This thesis can be read in two ways: as a claim about what structures refer to, or about how we conceptualize them. Objects-modelism, developed by Button and Walsh, pursues the first; the second leads to concepts-modelism, which remains underexplored. In this paper we develop and defend a version of concepts-modelism, cognitive modelism, drawing on Carey’s theory of conceptual development, and we show how it addresses the challenges Button and Walsh pose for a conceptual account of mathematical structures.
Read moreUnderstanding mental health discourse on Reddit with transformers and explainability.
Social media is a powerful tool also for discussing mental health. The conversations that take place in these spaces provide a unique insight into how users talk about the issue. This study uses fine-tuned pretrained transformer models (BERT and MentalBERT), to classify Reddit posts about anxiety, depression, bipolar disorder and borderline personality disorder (BPD) in specialised subreddits. By assessing how well subreddit conversations align with their intended mental health focus, the analysis ensures that these communities are effectively serving their purpose as support spaces. Our classification models achieve an average accuracy of 82%, with MentalBERT slightly outperforming BERT. To ensure transparency, we use Local Interpretable Model-agnostic Explanations (LIME) to identify key linguistic patterns that influence the model predictions. The outcome reveals distinct language use across conditions: as examples, discussions in bipolar disorder subreddits often refer to mood instability, while BPD communities emphasise challenges in emotional regulation. By integrating classification with explainability, this study offers insights into thematic patterns in online discourse that can support mental health professionals in identifying trends. While our models are not diagnostic tools, they function as subreddit-alignment classifiers, helping to uncover how different topics are discussed across communities. These insights may inform human-in-the-loop community management strategies and contribute to raising awareness and reducing stigma around mental health issues, ultimately fostering more supportive digital environments.
Read moreBusiness groups as knowledge-based hierarchies of firms
Abstract We provide the first worldwide overview of the patterns of hierarchical differentiation across Business Groups (BGs), highlighting the coexistence of different hierarchical shapes. We show how the different shapes can arise as optimal hierarchical structures in a knowledge-based model of BGs when subsidiaries’ operations involve problem-solving under parents’ supervision. The optimal choice of hierarchical structure is driven by production efficiency and two dimensions of problem-solving: efficiency related to supervising knowledge creation and handling associated communication across subsidiaries. We check the consistency of the model’s predictions with the empirical patterns. The model successfully passes the consistency test (JEL D23, L23, F23, L25, G34)
Read moreA method for real-world privacy-preserving Android malware detection through Federated Machine Learning
Privacy is one of the most critical issues associated with spreading the Internet of Things and Internet of Everything devices. Over the years, several methods have been introduced to address this phenomenon. In 2017, Google introduced the concept of Federated Machine Learning. This paradigm allows models to be trained collaboratively across multiple decentralized devices or servers, holding local data samples without exchanging them. This approach enhances data privacy and security by ensuring raw data remains on local devices while only model updates are shared and aggregated. This paper presents a privacy-preserving Android malware detector based on Federated Machine Learning. As a first step, we built a dataset comprising over 40,000 Android applications, including trusted and malicious (belonging to 71 malware families) samples. Afterward, we conducted experiments leveraging three different architectures by exploiting the CIFAR-10 and the ImageNet datasets, employing hyperparameters determined through a Grid Search algorithm by exploiting 40 clients. Moreover, the experimental analysis uses two distributions: Independent and identically distributed and non-independent and identically distributed data. To conclude the Federated Machine Learning experiments, we trained models for each architecture, with both weight types and distribution models, by applying the Clipping Norm Aggregator. The results exhibit interesting performances with Independent and identically distributed data, achieving an accuracy of 0.873 without normalization and 0.877 with the Clipping Norm aggregator. However, with non-independent and identically distributed data, the model accuracy equals 0.865 without normalization, 0.864 with the Clipping Norm aggregator using Custom MobileNet 2. In conclusion, to compare Federated Machine Learning with a centralized training approach, we trained several models adopting the same dataset, dataset splitting, and architectures, achieving an accuracy of 0.944 using InceptionV3. The outcomes show that the proposed method can provide engaging performances in privacy-preserving Android malware detection.
Read more