Giovanna Maria Dimitri

Cultural heritage plays a crucial role for societal identity, allowing a deep connection between individuals and their historical roots. In this context the advent of deep learning technologies has significantly influenced the preservation and analysis of cultural heritage, offering innovative and technological solutions for heritage documentation, classification, and restoration. This article presents a short review of existing deep learning methodologies applied to cultural heritage preservation in Italy. By leveraging a systematic literature search, we investigate technological solutions deployed across various Italian regions and analyze their impact on the dissemination of artificial intelligence in this field. Furthermore, we perform a geolocation-based assessment of these approaches to uncover potential correlations between technological novel solutions and regional characteristics. Our findings indicate the critical role of deep learning in modernizing heritage conservation efforts, moreover, highlighting the necessity for continued research and interdisciplinary collaboration in this domain.

The use of social networks has been shown as a powerful tool for driving populations opinions towards specific trends and interests. Yet what actually makes the success of a profile? Are emotions responsible for driving the public opinion and the opinion of the followers? We present a study on the influence of emotions in their success. To do so, we first created a novel dataset called InfluEmo, crawled from Instagram, in which we designed and analyzed the impact of emotions in influencers’ success. The dataset InfluEmo is novel and freely available. Automatic emotion extraction yielded promising results, supporting our hypothesis that specific emotional profiles in influencers’ posted content are associated with measurable indicators of success measured as number of followers. These findings suggest that emotions might play a systematic and quantifiable role in shaping public opinion and influencing users’ interactions on Instagram. Using the novel InfluEmo dataset (≈38,000 posts, ≈970 profiles, 4 domains: fashion, climate, AI, and journalism), the paper shows, in fact, that more positive emotional language is consistently associated with higher engagement, with fashion influencers achieving the highest average likes (≈138,885/post) and lowest emotional entropy, while AI, climate, and journalism content—characterized by more neutral or mixed emotions—exhibits lower likes (≈6761–19,544/post), weaker sentiment–likes correlations, and higher entropy, indicating that positivity and emotional predictability outperform informational complexity in driving Instagram success.

Large Language Models (LLMs) promise to enhance clinical decision-making, yet empirical studies reveal a paradox: physician performance with LLM assistance shows minimal improvement or even deterioration. This failure stems from an “acquiescence problem”: current LLMs passively confirm rather than challenge clinicians’ hypotheses, reinforcing cognitive biases such as anchoring and premature closure. To address these limitations, we propose a Dialogic Reasoning Framework that operationalizes Dialogical AI principles through a prototype implementation named “Diagnostic Dialogue” (DiDi). This framework operationalizes LLMs into three user-controlled roles: the Framework Coach (guiding structured reasoning), the Socratic Guide (asking probing questions), and the Red Team Partner (presenting evidence-based alternatives). Built upon Retrieval-Augmented Generation (RAG) architecture for factual grounding and traceability, this framework transforms LLMs from passive information providers into active reasoning partners that systematically mitigate cognitive bias. We evaluate the feasibility and qualitative impact of this framework through a pilot study (DiDi) deployed at Centro Chirurgico Toscano (CCT). Through purposive sampling of complex clinical scenarios, we present comparative case studies illustrating how the dialogic approach generates necessary cognitive friction to overcome acquiescence observed in standard LLM interactions. While rigorous clinical validation through randomized controlled trials remains necessary, this work establishes a methodological foundation for designing LLM-based clinical decision support systems that genuinely augment human clinical reasoning.

Continual learning (CL) has emerged as an important machine learning paradigm for developing adaptive artificial intelligence systems capable of life long learning, while addressing neural network challenges such as catastrophic forgetting. In this article, we provide a systematic review of the applications of CL in the agrifood sector. We organize our review by first introducing the fundamental principles of CL and specifying the methodology adopted for the definition of the relevant queries, the retrieval of the papers in the selected sources and the analysis of the literature. We then present and describe existing CL approaches applied to agrifood systems, identifying trends and challenges. Through our extensive literature survey, we highlight main characteristics of CL approaches in the context of agriculture, discussing current views and future perspectives.

Global Navigation Satellite System (GNSS) meteorology has emerged as a valuable tool for atmospheric monitoring, providing high-resolution, near-real-time data that can significantly improve precipitation nowcasting. This study aims to enhance short-term precipitation forecasting by integrating GNSS-derived Precipitable Water Vapor (PWV)—a key indicator of atmospheric moisture—with traditional meteorological observations. A novel two-step machine learning framework is proposed that combines a Random Forest (RF) model and a Long Short-Term Memory (LSTM) neural network. The RF model first estimates current precipitation based on PWV, surface weather parameters, and auxiliary atmospheric variables. Then, the LSTM network leverages temporal dependencies within the data to predict precipitation for the subsequent hour. This hybrid method capitalizes on the RF’s ability to model complex nonlinear relationships and the LSTM’s strength in handling time series data. The results demonstrate that the proposed approach improves forecasting accuracy, particularly during extreme weather events such as intense rainfall and thunderstorms, outperforming conventional models. By integrating GNSS meteorology with advanced machine learning techniques, this study offers a promising tool for meteorological services, early warning systems, and disaster risk management. The findings highlight the potential of GNSS-based nowcasting for real-time decision-making in weather-sensitive applications.

Traceability plays a critical role in ensuring the quality, safety, and transparency of supply chains, where transportation stakeholders are fundamental to the efficient movement of goods. However, the diversity of actors involved poses significant challenges to achieving these goals. Each organization typically operates its own information system, tailored to manage internal data, but often lacks the ability to communicate effectively with external systems. Moreover, when data exchange between different systems is required, it becomes critical to maintain full control over the shared data and to manage access rights precisely. In this work, we propose the concept of interoperable traceability. We present a model that enables the seamless integration of data from sensors, IoT devices, data management platforms, and distributed ledger technologies (DLT) within a newly designed data space architecture. We also demonstrate a practical implementation of this concept by applying it to real-world scenarios in the agri-food sector, with direct implications for transportation systems and all stakeholders in a supply chain. Our demonstrator supports the secure exchange of traceability data between existing systems, providing stakeholders with a novel approach to managing and auditing data with increased transparency and efficiency.

Despite the growing popularity of organic foods, research on their effects on human health, particularly regarding cancer and diabetes, remains limited. While some studies suggest potential health benefits, others yield conflicting results or lack sufficient evidence to draw conclusions. Understanding the causal relationship between organic food consumption and health outcomes is challenging, especially with limited datasets. Our study examines the correlation between organic food consumption and the prevalence of cancer and diabetes in European nations over time. We compared these findings with data from 100 Italian citizens regarding their perceptions of organic food’s health benefits collected through a novel questionnaire. To identify patterns, we applied Affinity Propagation clustering to group countries based on organic food consumption and disease prevalence. We also created an animated map to visualize cluster progression over time and used the Global Multiplexity Index to evaluate consistency. Our analysis revealed two subgroups of European countries exhibiting significant similarities in organic food consumption and health outcomes. The clustering analysis performed year-by-year on three variables across European nations using the Affinity Propagation algorithm revealed that two clusters consistently maximized the Global Multiplexity Index over time. The first cluster included Belgium, Finland, Ireland, Italy, and Spain, while the second comprised Bulgaria, Turkey, Romania, Ukraine, Czech Republic, Hungary, Poland, Greece, and Russia. These clusters displayed distinct trends concerning sustainable development goals (SDGs) related to organic farming and non-communicable diseases. Additionally, mapping SDG indicators along with geographic and socio-economic factors supported our findings. Moreover, we introduced a novel dataset and offered insights into both the European context and the Italian scenario, contributing to further research on organic food’s impact on public health.