Typology
operator:
and / or
Language
operator:
and / or
Date
operator:
and / or
Rights
operator:
and / or
2026
Journal article
Open Access
Enhancing token boundary detection in disfluent speech
Srivastava Manu, Ferro Marcello, Pirrelli Vito, Coro Gianpaolo
This paper presents an open-source Automatic Speech Recognition (ASR) pipeline optimised for disfluent Italian read speech, designed to enhance both transcription accuracy and token boundary precision in low-resource settings. The study aims to address the difficulty that conventional ASR systems face in capturing the temporal irregularities of disfluent reading, which are crucial for psycholinguistic and clinical analyses of fluency. Building upon the WhisperX framework, the proposed system replaces the neural Voice Activity Detection module with an energy-based segmentation algorithm designed to preserve prosodic cues such as pauses and hesitations. A dual-alignment strategy integrates two complementary phoneme-level ASR models to correct onset–offset asymmetries, while a bias-compensation post-processing step mitigates systematic timing errors. Evaluation on the READLET (child read speech) and CLIPS (adult read speech) corpora shows consistent improvements over baseline systems, confirming enhanced robustness in boundary detection and transcription under disfluent conditions. The results demonstrate that the proposed architecture provides a general, language-independent framework for accurate alignment and disfluency-aware ASR. The approach can support downstream analyses of reading fluency and speech planning, contributing to both computational linguistics and clinical speech research.Source: INTELLIGENT SYSTEMS WITH APPLICATIONS, vol. 29
DOI: 10.1016/j.iswa.2025.200614
Project(s): READLET
Metrics:
Srivastava Manu, Ferro Marcello, Pirrelli Vito, Coro Gianpaolo
This paper presents an open-source Automatic Speech Recognition (ASR) pipeline optimised for disfluent Italian read speech, designed to enhance both transcription accuracy and token boundary precision in low-resource settings. The study aims to address the difficulty that conventional ASR systems face in capturing the temporal irregularities of disfluent reading, which are crucial for psycholinguistic and clinical analyses of fluency. Building upon the WhisperX framework, the proposed system replaces the neural Voice Activity Detection module with an energy-based segmentation algorithm designed to preserve prosodic cues such as pauses and hesitations. A dual-alignment strategy integrates two complementary phoneme-level ASR models to correct onset–offset asymmetries, while a bias-compensation post-processing step mitigates systematic timing errors. Evaluation on the READLET (child read speech) and CLIPS (adult read speech) corpora shows consistent improvements over baseline systems, confirming enhanced robustness in boundary detection and transcription under disfluent conditions. The results demonstrate that the proposed architecture provides a general, language-independent framework for accurate alignment and disfluency-aware ASR. The approach can support downstream analyses of reading fluency and speech planning, contributing to both computational linguistics and clinical speech research.Source: INTELLIGENT SYSTEMS WITH APPLICATIONS, vol. 29
DOI: 10.1016/j.iswa.2025.200614
Project(s): READLET
Metrics:
See at:
CNR IRIS 
| www.sciencedirect.com | CNR IRIS 
2026
Journal article
Open Access
Detecting fishing areas from navigation radar detector data
Coro Gianpaolo, Bove Pasquale, Ellenbroek Anton
Detecting global fishing activity is essential for sustainable ocean governance, yet systems based on vessel-transmitted information, such as Automatic Identification System (AIS) and Vessel Monitoring Systems, are limited by access issues, coverage gaps, and the inability to detect non-cooperative vessels. To overcome these issues, this paper presents Point-to-Fishing (P2F), an AI-driven workflow to detect fishing areas and estimate fishing hours from Navigation Radar Detector (NRD) data of satellite or terrestrial systems, complemented with currents and bathymetry data from Copernicus and GEBCO. P2F integrates analytical components based on statistical analysis, machine learning, and deep learning to conduct vessel behaviour analysis, spatial feature extraction (vessel abundance, recurrence, current-driven interpolation, and bathymetric suitability), and anomaly detection. The workflow operates effectively with or without vessel identifiers, enabling the detection of fishing areas in data-sparse or AIS-denied regions, even using one satellite only. P2F is validated on data covering the North Sea, the Western Norwegian Sea, and the North Atlantic. The validation cases utilise terrestrial and satellite NRD data alternately, with the Global Fishing Watch fishing effort distributions as a validation reference. P2F achieves a consistent ~75% agreement in relevant fishing area classification and intense-fishing area identification, and ~93% accuracy in total fishing effort estimation.Source: INTERNATIONAL JOURNAL OF DIGITAL EARTH, vol. 19 (issue 1)
DOI: 10.1080/17538947.2026.2617004
DOI: https://doi.org/10.1080/17538947.2026.2617004
Project(s): Accordo di Collaborazione N. 403036 del 25/10/2024 tra la Norwegian Defence Research Establishment (FFI) e il CNR-ISTI.
Metrics:
Coro Gianpaolo, Bove Pasquale, Ellenbroek Anton
Detecting global fishing activity is essential for sustainable ocean governance, yet systems based on vessel-transmitted information, such as Automatic Identification System (AIS) and Vessel Monitoring Systems, are limited by access issues, coverage gaps, and the inability to detect non-cooperative vessels. To overcome these issues, this paper presents Point-to-Fishing (P2F), an AI-driven workflow to detect fishing areas and estimate fishing hours from Navigation Radar Detector (NRD) data of satellite or terrestrial systems, complemented with currents and bathymetry data from Copernicus and GEBCO. P2F integrates analytical components based on statistical analysis, machine learning, and deep learning to conduct vessel behaviour analysis, spatial feature extraction (vessel abundance, recurrence, current-driven interpolation, and bathymetric suitability), and anomaly detection. The workflow operates effectively with or without vessel identifiers, enabling the detection of fishing areas in data-sparse or AIS-denied regions, even using one satellite only. P2F is validated on data covering the North Sea, the Western Norwegian Sea, and the North Atlantic. The validation cases utilise terrestrial and satellite NRD data alternately, with the Global Fishing Watch fishing effort distributions as a validation reference. P2F achieves a consistent ~75% agreement in relevant fishing area classification and intense-fishing area identification, and ~93% accuracy in total fishing effort estimation.Source: INTERNATIONAL JOURNAL OF DIGITAL EARTH, vol. 19 (issue 1)
DOI: 10.1080/17538947.2026.2617004
DOI: https://doi.org/10.1080/17538947.2026.2617004
Project(s): Accordo di Collaborazione N. 403036 del 25/10/2024 tra la Norwegian Defence Research Establishment (FFI) e il CNR-ISTI.
Metrics:
2026
Conference article
Open Access
A conversational assistant for geoscientists in virtual research environments
Peccerillo Biagio, Oliviero Alfredo, Procaccini Marco, Candela Leonardo, Frosini Luca, Mangiacrapa Francesco, Panichi Giancarlo, Assante Massimiliano, Pagano Pasquale
D4Science provides web-based Virtual Research Environments (VREs) that support FAIR, open, and reproducible science across multiple research domains, including Earth science. These environments integrate data access, computation, and collaboration services, offering powerful capabilities to researchers and enabling complex, data-intensive scientific activities within a shared digital infrastructure. This contribution introduces a conversational intelligent assistant integrated into D4Science VREs, designed to support Earth scientists in their research activity. The assistant provides a natural language interface that helps users interact with D4Science VREs' services, locate relevant datasets and research items, obtain guidance on common tasks, and support exploratory and operational activities within the VRE. The assistant is designed with a modular approach. The user interacts with a coordinator agent that orchestrates a multi-agent system, where specialized AI agents collaborate to perform a variety of tasks. This architecture allows the assistant to handle heterogeneous requests and to support users across different phases of their research activities, while also facilitating maintenance and extensibility. The conversational agent adopts a Retrieval-Augmented Generation (RAG) approach that leverages the knowledge already captured by the VRE through its regular use by research communities. In fact, as VREs naturally accumulate updated knowledge created and curated by researchers over time, the assistant's knowledge base evolves incorporating new information. This way, the assistant can ground its responses in domain-specific and up-to-date information, effectively acting as a domain-aware expert embedded within the research environment. By serving as an accessible entry point to the VRE, the assistant complements existing interfaces without altering established workflows. The presentation discusses the motivation, design choices, and integration strategy. It also presents various concrete use cases relevant to Earth scientists, demonstrating how the conversational assistant can be effectively employed to support their research activity.DOI: 10.5194/egusphere-egu26-13138
Metrics:
Peccerillo Biagio, Oliviero Alfredo, Procaccini Marco, Candela Leonardo, Frosini Luca, Mangiacrapa Francesco, Panichi Giancarlo, Assante Massimiliano, Pagano Pasquale
D4Science provides web-based Virtual Research Environments (VREs) that support FAIR, open, and reproducible science across multiple research domains, including Earth science. These environments integrate data access, computation, and collaboration services, offering powerful capabilities to researchers and enabling complex, data-intensive scientific activities within a shared digital infrastructure. This contribution introduces a conversational intelligent assistant integrated into D4Science VREs, designed to support Earth scientists in their research activity. The assistant provides a natural language interface that helps users interact with D4Science VREs' services, locate relevant datasets and research items, obtain guidance on common tasks, and support exploratory and operational activities within the VRE. The assistant is designed with a modular approach. The user interacts with a coordinator agent that orchestrates a multi-agent system, where specialized AI agents collaborate to perform a variety of tasks. This architecture allows the assistant to handle heterogeneous requests and to support users across different phases of their research activities, while also facilitating maintenance and extensibility. The conversational agent adopts a Retrieval-Augmented Generation (RAG) approach that leverages the knowledge already captured by the VRE through its regular use by research communities. In fact, as VREs naturally accumulate updated knowledge created and curated by researchers over time, the assistant's knowledge base evolves incorporating new information. This way, the assistant can ground its responses in domain-specific and up-to-date information, effectively acting as a domain-aware expert embedded within the research environment. By serving as an accessible entry point to the VRE, the assistant complements existing interfaces without altering established workflows. The presentation discusses the motivation, design choices, and integration strategy. It also presents various concrete use cases relevant to Earth scientists, demonstrating how the conversational assistant can be effectively employed to support their research activity.DOI: 10.5194/egusphere-egu26-13138
Metrics:
See at:
CNR IRIS | www.egu26.eu | CNR IRIS
2026
Journal article
Open Access
The OpenAIRE Graph: enabling open research intelligence using open data
Manghi Paolo
The OpenAIRE Graph provides an open, community-governed data infrastructure for research intelligence, enabling transparent and auditable use of scholarly data beyond proprietary systems. The global research ecosystem is calling for a structural transition from proprietary, opaque systems for research intelligence to open, community-governed infrastructures. At the centre of this shift is the need to reclaim how scholarly data is collected, connected, and used to inform research evaluation and policy. The OpenAIRE Graph addresses this challenge by providing a large-scale, openly accessible scholarly knowledge graph that treats publications, data, and software as first-class research outputs. As a community-governed infrastructure, it establishes a transparent and auditable foundation for Open Research Intelligence.Source: ERCIM NEWS, vol. 144, pp. 8-9
Manghi Paolo
The OpenAIRE Graph provides an open, community-governed data infrastructure for research intelligence, enabling transparent and auditable use of scholarly data beyond proprietary systems. The global research ecosystem is calling for a structural transition from proprietary, opaque systems for research intelligence to open, community-governed infrastructures. At the centre of this shift is the need to reclaim how scholarly data is collected, connected, and used to inform research evaluation and policy. The OpenAIRE Graph addresses this challenge by providing a large-scale, openly accessible scholarly knowledge graph that treats publications, data, and software as first-class research outputs. As a community-governed infrastructure, it establishes a transparent and auditable foundation for Open Research Intelligence.Source: ERCIM NEWS, vol. 144, pp. 8-9
See at:
ercim-news.ercim.eu | CNR IRIS | CNR IRIS
2026
Journal article
Open Access
Charting the landscape of Italian Diamond Open Access publishing
Angioni Simone, Baglioni Miriam, Bardi Alessia, Manghi Paolo, Mannocci Andrea, Pavone Gina
Diamond Open Access (DOA) is a non-commercial model of scholarly publishing that removes financial barriers for authors and readers. While international studies have outlined the global uptake of DOA, this paper investigates the presence and characteristics of the DOA landscape in Italy. We conducted a quantitative analysis on the 168 Italian journals classified as Diamond by EZB, and we studied their publishing volume, disciplinary distribution and citation impact by integrating information from the following open resources: DOAJ, OpenAIRE, ROAD, SCImago Journal Rank, and the ANVUR classification used in the Italian Research Assessment Framework (VQR). Key findings include the significant growth of DOA journals, particularly in the social sciences and humanities, as well as the high level of international citations, indicating strong global relevance. The study also highlights challenges such as the need for better indexing and comprehensive data to fully capture the DOA landscape.Source: QUANTITATIVE SCIENCE STUDIES
DOI: 10.1162/qss.a.466
Metrics:
Angioni Simone, Baglioni Miriam, Bardi Alessia, Manghi Paolo, Mannocci Andrea, Pavone Gina
Diamond Open Access (DOA) is a non-commercial model of scholarly publishing that removes financial barriers for authors and readers. While international studies have outlined the global uptake of DOA, this paper investigates the presence and characteristics of the DOA landscape in Italy. We conducted a quantitative analysis on the 168 Italian journals classified as Diamond by EZB, and we studied their publishing volume, disciplinary distribution and citation impact by integrating information from the following open resources: DOAJ, OpenAIRE, ROAD, SCImago Journal Rank, and the ANVUR classification used in the Italian Research Assessment Framework (VQR). Key findings include the significant growth of DOA journals, particularly in the social sciences and humanities, as well as the high level of international citations, indicating strong global relevance. The study also highlights challenges such as the need for better indexing and comprehensive data to fully capture the DOA landscape.Source: QUANTITATIVE SCIENCE STUDIES
DOI: 10.1162/qss.a.466
Metrics:
See at:
direct.mit.edu | CNR IRIS | CNR IRIS
2026
Journal article
Open Access
Belief change based on knowledge measures
Casini Giovanni, Straccia Umberto
Belief Change (BC) is the process of changing beliefs (in our case, in terms of contraction, expansion and revision) taking into account a new piece of knowledge, which possibly may be in contradiction with the current belief. In this work, we present a novel quantitative BC framework based on the principle of minimizing the surprise from an information-theoretic perspective. Central to our approach is the Principle of Minimal Surprise (PMS), which asserts that when confronted with the uncertainty about which is the actual world, an agent should tend to favour the most expected, i.e., least surprising, outcomes. To formalize this, we make use of Knowledge Measures (KMs), which quantify the amount of information contained in a knowledge base. Guided by the PMS, our framework encourages belief change operations that minimize the informational amount deviation from the original belief, i.e., those that introduce the least surprise.Source: JOURNAL OF LOGIC AND COMPUTATION, vol. 36 (issue 2)
DOI: 10.1093/logcom/exag007
Project(s): STARWARS
Metrics:
Casini Giovanni, Straccia Umberto
Belief Change (BC) is the process of changing beliefs (in our case, in terms of contraction, expansion and revision) taking into account a new piece of knowledge, which possibly may be in contradiction with the current belief. In this work, we present a novel quantitative BC framework based on the principle of minimizing the surprise from an information-theoretic perspective. Central to our approach is the Principle of Minimal Surprise (PMS), which asserts that when confronted with the uncertainty about which is the actual world, an agent should tend to favour the most expected, i.e., least surprising, outcomes. To formalize this, we make use of Knowledge Measures (KMs), which quantify the amount of information contained in a knowledge base. Guided by the PMS, our framework encourages belief change operations that minimize the informational amount deviation from the original belief, i.e., those that introduce the least surprise.Source: JOURNAL OF LOGIC AND COMPUTATION, vol. 36 (issue 2)
DOI: 10.1093/logcom/exag007
Project(s): STARWARS
Metrics:
See at:
academic.oup.com | CNR IRIS | CNR IRIS
2026
Book
Open Access
Advancing Open Science: federated infrastructures and trustworthy ecosystems
Candela Leonardo, Di Cosmo Roberto
Open Science is a broad and evolving movement. The UNESCO framework describes it as an inclusive approach aimed at making scientific knowledge openly available, accessible, and reusable for everyone, opening the processes of knowledge creation and evaluation to stakeholders beyond the traditional research community [1]. Today, Open Science is no longer merely a normative ideal: it has become an operational requirement embedded in national strategies, funding conditions, and research assessment reforms across Europe and beyond. Yet the very breadth of Open Science is also its greatest challenge. The movement rests on several distinct pillars, each with its own history, infrastructure landscape, and degree of maturity — and each exposed to the same structural risk: fragmentation. The three pillars - and the fragmentation trap The oldest pillar is open access to publications. Decades of effort have produced undeniable progress, but also a cautionary tale. Because coordination came late, the landscape is now highly fragmented: OpenDOAR counts over 6,000 open access repositories worldwide, each requiring its own infrastructure, archival, backup, and metadata curation. Content is duplicated, metadata is inconsistent, and the cost of maintaining this patchwork is borne many times over. The recent move to fund, via national grants, the EU-originated Open Research Europe journal illustrates how difficult it is to retrofit coherence onto an ecosystem that grew without a shared architectural plan. The second pillar, open research data, has benefited from the lessons of publications and from the early adoption of the FAIR principles. Yet a similar proliferation of platforms and curation challenges is already visible, with a very long tail of research data that struggles to find a sustainable home. National initiatives such as Recherche Data Gouv in France and the PLATICA project in Spain point toward a promising model: shared, mutualized infrastructures that host curated research data as a public good, rather than leaving each institution to build and maintain its own silo. The third pillar — research software — has long pre-existed the others, since software has been at the heart of scientific computation for decades. Yet it was recognized as a pillar of Open Science only very recently. The French Second National Plan for Open Science (2021) was the first national strategy to dedicate a full chapter to software, establishing measures for archiving, referencing, and citing source code, creating a national research software award, and providing explicit support for Software Heritage as a key infrastructure [2]. Spain is now actively building on this momentum, as evidenced by the discussions at the recent second national days on Open Science held in Aranjuez in March 2026. For software, there is a unique opportunity to avoid the fragmentation that has plagued publications and data. Software Heritage was designed from the outset as a universal, open, non-profit archive for all software source code. It already preserves over 28 billion source files from more than 430 million projects collected across over 5,000 code hosting and distribution platforms worldwide, assigning intrinsic, cryptographically strong identifiers (SWHIDs, now standardized as ISO/IEC 18670). This provides a single, shared layer for archiving, referencing, describing, and citing software — a foundation that Open Science policy can build on directly, without the need to reconcile thousands of independent local repositories after the fact. Federating from the top: promise and friction Alongside bottom-up infrastructure efforts, Europe has invested heavily in top-down coordination through the European Open Science Cloud (EOSC), which aims to federate existing services into an interoperable, cross-border research environment. Several contributions to this issue illustrate both the promise and the complexity of this endeavour. Yet federation by decree is hard. Even in countries with active EOSC engagement, surveys show that a majority of researchers still store data primarily on personal computers, and awareness of federated infrastructure remains low. The gap between policy ambition and daily research practice is real, and bridging it requires not just technical platforms but sustained investment in skills, incentives, and institutional culture change. A map of the current landscape The contributions collected in this special theme offer a cross-section of the current European effort, organised into five thematic clusters. A first cluster addresses research assessment and scholarly representation. The OpenAIRE Graph (Manghi) provides a community-governed scholarly knowledge graph treating datasets and software as first-class outputs, offering an open alternative to proprietary research intelligence. MyResearchFolio (Amodeo and Xenou) builds on this to support richer researcher profiles aligned with responsible assessment principles, while BibTexViz (Horcas) demonstrates visual analytics for open bibliographic data. The EOSC Open Science Observatory (Szybisty) combines indicators, national narratives, and AI-assisted analysis to monitor Open Science progress across Europe. A second cluster explores the transition from FAIR data to AI-ready workflows. Contributions show how shared industrial datasets can feed collaborative knowledge pipelines (Gorissen and Brauner), how compute-to-data architectures enable scalable analysis on research infrastructures (Brus et al.), and how modular, open-source research software frameworks can support advanced biomedical analytics (Segura-Ortiz et al.). A third cluster highlights semantic foundations and knowledge graph infrastructures as critical enablers of interoperability, through the transformation of legacy databases into FAIR-by-design knowledge graphs (Marketakis et al.) and the evolution of the EOSC Interoperability Framework toward machine-actionable, composable service templates (Bardi et al.). A fourth cluster addresses the governance, skills, sovereignty, and ethical foundations without which technical infrastructure cannot function. Contributions cover human-centred threat modelling (Onofri and Corti), structured co-creation in data spaces (Stampfl and Palkovits-Rauter), Open Science education beyond purely technical skills (Flicker et al.), the Czech national experience with FAIR adoption (Dvořák et al.), the tension between Creative Commons licences and AI training (Spichtinger), and privacy-enhancing technologies for secure cross-border data sharing (Jimenez-Bejarano et al.). The fifth and final cluster presents operational experiences with federated science gateways, including the EOSC EU Node (Brunschweiger et al.), the Innovation Sandbox (Drago and Fiore), the Data Commons (Fernández and Fava), the ENVRI-Hub for environmental research (Drago et al.), the D4Science virtual research environments (Assante et al.), and the DAVE conversational AI assistant for navigating complex research workflows (Dell'Amico et al.). Looking ahead Taken together, these contributions make clear that the next phase of Open Science will be defined not just by openness, but by trustworthiness and integration. Several priorities stand out. First, avoiding fragmentation must become a conscious design principle, not an afterthought. For each pillar of Open Science — publications, data, and software — the question is whether we build shared, mutualized infrastructure from the start or spend decades trying to harmonize a patchwork. Second, research assessment must formally recognize the full range of research outputs — datasets, software, workflows — alongside publications, moving away from proprietary metrics toward transparent, community-governed research intelligence. Third, the intersection of open licensing and AI training remains legally ambiguous. As AI models increasingly consume open research data and code, robust opt-in/opt-out mechanisms and legal clarity are urgently needed. Finally, long-term financial sustainability for community-governed infrastructure remains an open problem. Short-term project funding cannot secure the digital commons on which European research increasingly depends. If the first phase of Open Science was about making research outputs accessible, the present phase is about making research ecosystems interoperable, intelligent, and trustworthy. The contributions in this issue offer both concrete experiences and forward-looking perspectives on how Europe is working to make that vision a reality.Source: ERCIM NEWS, vol. 144, pp. 6-40
Candela Leonardo, Di Cosmo Roberto
Open Science is a broad and evolving movement. The UNESCO framework describes it as an inclusive approach aimed at making scientific knowledge openly available, accessible, and reusable for everyone, opening the processes of knowledge creation and evaluation to stakeholders beyond the traditional research community [1]. Today, Open Science is no longer merely a normative ideal: it has become an operational requirement embedded in national strategies, funding conditions, and research assessment reforms across Europe and beyond. Yet the very breadth of Open Science is also its greatest challenge. The movement rests on several distinct pillars, each with its own history, infrastructure landscape, and degree of maturity — and each exposed to the same structural risk: fragmentation. The three pillars - and the fragmentation trap The oldest pillar is open access to publications. Decades of effort have produced undeniable progress, but also a cautionary tale. Because coordination came late, the landscape is now highly fragmented: OpenDOAR counts over 6,000 open access repositories worldwide, each requiring its own infrastructure, archival, backup, and metadata curation. Content is duplicated, metadata is inconsistent, and the cost of maintaining this patchwork is borne many times over. The recent move to fund, via national grants, the EU-originated Open Research Europe journal illustrates how difficult it is to retrofit coherence onto an ecosystem that grew without a shared architectural plan. The second pillar, open research data, has benefited from the lessons of publications and from the early adoption of the FAIR principles. Yet a similar proliferation of platforms and curation challenges is already visible, with a very long tail of research data that struggles to find a sustainable home. National initiatives such as Recherche Data Gouv in France and the PLATICA project in Spain point toward a promising model: shared, mutualized infrastructures that host curated research data as a public good, rather than leaving each institution to build and maintain its own silo. The third pillar — research software — has long pre-existed the others, since software has been at the heart of scientific computation for decades. Yet it was recognized as a pillar of Open Science only very recently. The French Second National Plan for Open Science (2021) was the first national strategy to dedicate a full chapter to software, establishing measures for archiving, referencing, and citing source code, creating a national research software award, and providing explicit support for Software Heritage as a key infrastructure [2]. Spain is now actively building on this momentum, as evidenced by the discussions at the recent second national days on Open Science held in Aranjuez in March 2026. For software, there is a unique opportunity to avoid the fragmentation that has plagued publications and data. Software Heritage was designed from the outset as a universal, open, non-profit archive for all software source code. It already preserves over 28 billion source files from more than 430 million projects collected across over 5,000 code hosting and distribution platforms worldwide, assigning intrinsic, cryptographically strong identifiers (SWHIDs, now standardized as ISO/IEC 18670). This provides a single, shared layer for archiving, referencing, describing, and citing software — a foundation that Open Science policy can build on directly, without the need to reconcile thousands of independent local repositories after the fact. Federating from the top: promise and friction Alongside bottom-up infrastructure efforts, Europe has invested heavily in top-down coordination through the European Open Science Cloud (EOSC), which aims to federate existing services into an interoperable, cross-border research environment. Several contributions to this issue illustrate both the promise and the complexity of this endeavour. Yet federation by decree is hard. Even in countries with active EOSC engagement, surveys show that a majority of researchers still store data primarily on personal computers, and awareness of federated infrastructure remains low. The gap between policy ambition and daily research practice is real, and bridging it requires not just technical platforms but sustained investment in skills, incentives, and institutional culture change. A map of the current landscape The contributions collected in this special theme offer a cross-section of the current European effort, organised into five thematic clusters. A first cluster addresses research assessment and scholarly representation. The OpenAIRE Graph (Manghi) provides a community-governed scholarly knowledge graph treating datasets and software as first-class outputs, offering an open alternative to proprietary research intelligence. MyResearchFolio (Amodeo and Xenou) builds on this to support richer researcher profiles aligned with responsible assessment principles, while BibTexViz (Horcas) demonstrates visual analytics for open bibliographic data. The EOSC Open Science Observatory (Szybisty) combines indicators, national narratives, and AI-assisted analysis to monitor Open Science progress across Europe. A second cluster explores the transition from FAIR data to AI-ready workflows. Contributions show how shared industrial datasets can feed collaborative knowledge pipelines (Gorissen and Brauner), how compute-to-data architectures enable scalable analysis on research infrastructures (Brus et al.), and how modular, open-source research software frameworks can support advanced biomedical analytics (Segura-Ortiz et al.). A third cluster highlights semantic foundations and knowledge graph infrastructures as critical enablers of interoperability, through the transformation of legacy databases into FAIR-by-design knowledge graphs (Marketakis et al.) and the evolution of the EOSC Interoperability Framework toward machine-actionable, composable service templates (Bardi et al.). A fourth cluster addresses the governance, skills, sovereignty, and ethical foundations without which technical infrastructure cannot function. Contributions cover human-centred threat modelling (Onofri and Corti), structured co-creation in data spaces (Stampfl and Palkovits-Rauter), Open Science education beyond purely technical skills (Flicker et al.), the Czech national experience with FAIR adoption (Dvořák et al.), the tension between Creative Commons licences and AI training (Spichtinger), and privacy-enhancing technologies for secure cross-border data sharing (Jimenez-Bejarano et al.). The fifth and final cluster presents operational experiences with federated science gateways, including the EOSC EU Node (Brunschweiger et al.), the Innovation Sandbox (Drago and Fiore), the Data Commons (Fernández and Fava), the ENVRI-Hub for environmental research (Drago et al.), the D4Science virtual research environments (Assante et al.), and the DAVE conversational AI assistant for navigating complex research workflows (Dell'Amico et al.). Looking ahead Taken together, these contributions make clear that the next phase of Open Science will be defined not just by openness, but by trustworthiness and integration. Several priorities stand out. First, avoiding fragmentation must become a conscious design principle, not an afterthought. For each pillar of Open Science — publications, data, and software — the question is whether we build shared, mutualized infrastructure from the start or spend decades trying to harmonize a patchwork. Second, research assessment must formally recognize the full range of research outputs — datasets, software, workflows — alongside publications, moving away from proprietary metrics toward transparent, community-governed research intelligence. Third, the intersection of open licensing and AI training remains legally ambiguous. As AI models increasingly consume open research data and code, robust opt-in/opt-out mechanisms and legal clarity are urgently needed. Finally, long-term financial sustainability for community-governed infrastructure remains an open problem. Short-term project funding cannot secure the digital commons on which European research increasingly depends. If the first phase of Open Science was about making research outputs accessible, the present phase is about making research ecosystems interoperable, intelligent, and trustworthy. The contributions in this issue offer both concrete experiences and forward-looking perspectives on how Europe is working to make that vision a reality.Source: ERCIM NEWS, vol. 144, pp. 6-40
See at:
ercim-news.ercim.eu | CNR IRIS | CNR IRIS
2026
Journal article
Open Access
Estimating species commonness and prevalence through unsupervised methods
Bove Pasquale, Bertini Andrea, Coro Gianpaolo
The prevalence of a species in a given area is crucial for estimating the environmental conditions associated with its subsistence within ecological niche models (ENMs). Prevalence is defined as the proportion of presences relative to the total number of sampled sites, reflecting prior expectation on species commonness or rarity. However, reliable estimation often faces challenges due to limited or biased occurrence data, particularly for rare or poorly monitored species. This work presents a data-driven, multi-species methodology to estimate species prevalence for use in ENMs. It leverages species occurrence records from the Global Biodiversity Information Facility and is entirely unsupervised. It utilises two clustering methods, one deep-learning model, and an ensemble model, plus statistical analysis to classify species commonness and transform classifications into prevalence probabilities. A case study is presented for 161 species living in the Massaciuccoli Lake basin (Tuscany, Italy), a wetland of high biodiversity value and ecological sensitivity. The models classified the species’ prevalence based on observations from other Italian wetland sites, and were evaluated against expert-based assessments. All models achieved high accuracy, with the deep-learning model achieving the highest (~ 81–90%). The proposed methodology is scalable and reproducible and can inform ENMs with objective, robust prevalence estimates.Source: SCIENTIFIC REPORTS, pp. 1-30
DOI: 10.1038/s41598-026-38900-1
DOI: https://doi.org/10.1038/s41598-026-38900-1
Project(s): Italian Integrated Environmental Research Infrastructures System
Metrics:
Bove Pasquale, Bertini Andrea, Coro Gianpaolo
The prevalence of a species in a given area is crucial for estimating the environmental conditions associated with its subsistence within ecological niche models (ENMs). Prevalence is defined as the proportion of presences relative to the total number of sampled sites, reflecting prior expectation on species commonness or rarity. However, reliable estimation often faces challenges due to limited or biased occurrence data, particularly for rare or poorly monitored species. This work presents a data-driven, multi-species methodology to estimate species prevalence for use in ENMs. It leverages species occurrence records from the Global Biodiversity Information Facility and is entirely unsupervised. It utilises two clustering methods, one deep-learning model, and an ensemble model, plus statistical analysis to classify species commonness and transform classifications into prevalence probabilities. A case study is presented for 161 species living in the Massaciuccoli Lake basin (Tuscany, Italy), a wetland of high biodiversity value and ecological sensitivity. The models classified the species’ prevalence based on observations from other Italian wetland sites, and were evaluated against expert-based assessments. All models achieved high accuracy, with the deep-learning model achieving the highest (~ 81–90%). The proposed methodology is scalable and reproducible and can inform ENMs with objective, robust prevalence estimates.Source: SCIENTIFIC REPORTS, pp. 1-30
DOI: 10.1038/s41598-026-38900-1
DOI: https://doi.org/10.1038/s41598-026-38900-1
Project(s): Italian Integrated Environmental Research Infrastructures System
Metrics:
2026
Journal article
Open Access
D4Science: an enabling infrastructure for open science
Assante Massimiliano, Frosini Luca, Mangiacrapa Francesco, Pagano Pasquale
D4Science supports Virtual Research Environments that integrate data, computing, and collaboration tools, making Open Science part of everyday research practice across diverse communities. Open Science is increasingly embedded in research practice, but implementing it in everyday workflows remains both a technological and organisational challenge. D4Science [1] addresses this challenge by adopting the as-a-Service paradigm and by offering Virtual Research Environments (VREs) [2], also called Virtual Laboratories (VLabs), as integrated, web-based, working environments. These environments provide researchers with seamless access to data, computational resources, and analytical services within a unified framework. By abstracting away the complexities of storage management, computation, and service orchestration, VREs enable scientists to focus on research design, methodological rigour and knowledge production rather than on IT and infrastructure concerns. However, fragmentation of tools and the effort required to prepare artefacts for reuse often hinder adoption. D4Science addresses this by embedding Open Science practices directly within VREs.Source: ERCIM NEWS, vol. 144, pp. 38-39
Assante Massimiliano, Frosini Luca, Mangiacrapa Francesco, Pagano Pasquale
D4Science supports Virtual Research Environments that integrate data, computing, and collaboration tools, making Open Science part of everyday research practice across diverse communities. Open Science is increasingly embedded in research practice, but implementing it in everyday workflows remains both a technological and organisational challenge. D4Science [1] addresses this challenge by adopting the as-a-Service paradigm and by offering Virtual Research Environments (VREs) [2], also called Virtual Laboratories (VLabs), as integrated, web-based, working environments. These environments provide researchers with seamless access to data, computational resources, and analytical services within a unified framework. By abstracting away the complexities of storage management, computation, and service orchestration, VREs enable scientists to focus on research design, methodological rigour and knowledge production rather than on IT and infrastructure concerns. However, fragmentation of tools and the effort required to prepare artefacts for reuse often hinder adoption. D4Science addresses this by embedding Open Science practices directly within VREs.Source: ERCIM NEWS, vol. 144, pp. 38-39
See at:
ercim-news.ercim.eu | CNR IRIS | CNR IRIS
2026
Journal article
Open Access
Interactive story maps for historical musical instruments: a 3D and semantic web tool for cultural heritage preservation
Bartalesi Valentina, Lenzi Emanuele, De Martino Claudio, Coro Gianpaolo
Ancient stringed instruments from the 17th and 18th centuries are essential for global cultural heritage but pose preservation and study challenges due to their fragility and rarity. Digital reproductions and interactive story maps offer new ways to analyse these instruments, providing historical, mechanical, and acoustic insights while aiding modern luthiers in replication. This paper presents a novel methodology and software that integrates 3D models of historical instruments into interactive story maps using Semantic Web technologies. The system enables users to explore 3D models with embedded annotations from Sketchfab and supports the creation of formal narratives enriched with geospatial and multimedia content. Each narrative event is linked to a backend knowledge base (KB) structured with established ontologies. The tool is open-source and web-based, and allows interoperability between its story maps and external KBs such as Wikidata and Europeana. It supports multiple levels of descriptive details, from precise 3D model annotations to broad geographic and temporal references. An evaluation case study featured 25 participants exploring a story map about a 1737 Antonio Stradivari violin, connected to a KB of 1716 triples. The results highlight the tool’s usability and effectiveness in representing the violin’s spatiotemporal and cultural context, showcasing the value of combining 3D models and geospatial data to enhance cultural heritage preservation.Source: INTERNATIONAL JOURNAL OF DATA SCIENCE AND ANALYTICS, vol. 22 (issue 1)
DOI: 10.1007/s41060-025-01014-4
Metrics:
Bartalesi Valentina, Lenzi Emanuele, De Martino Claudio, Coro Gianpaolo
Ancient stringed instruments from the 17th and 18th centuries are essential for global cultural heritage but pose preservation and study challenges due to their fragility and rarity. Digital reproductions and interactive story maps offer new ways to analyse these instruments, providing historical, mechanical, and acoustic insights while aiding modern luthiers in replication. This paper presents a novel methodology and software that integrates 3D models of historical instruments into interactive story maps using Semantic Web technologies. The system enables users to explore 3D models with embedded annotations from Sketchfab and supports the creation of formal narratives enriched with geospatial and multimedia content. Each narrative event is linked to a backend knowledge base (KB) structured with established ontologies. The tool is open-source and web-based, and allows interoperability between its story maps and external KBs such as Wikidata and Europeana. It supports multiple levels of descriptive details, from precise 3D model annotations to broad geographic and temporal references. An evaluation case study featured 25 participants exploring a story map about a 1737 Antonio Stradivari violin, connected to a KB of 1716 triples. The results highlight the tool’s usability and effectiveness in representing the violin’s spatiotemporal and cultural context, showcasing the value of combining 3D models and geospatial data to enhance cultural heritage preservation.Source: INTERNATIONAL JOURNAL OF DATA SCIENCE AND ANALYTICS, vol. 22 (issue 1)
DOI: 10.1007/s41060-025-01014-4
Metrics:
See at:
CNR IRIS | link.springer.com | CNR IRIS | CNR IRIS
2026
Journal article
Open Access
Does diversity of expertise drive citation impact? Evidence from Computer Science
Salatino Angelo, Osborne Francesco, Recupero Reforgiato Diego, Angioni Simone, Motta Enrico
High-quality scientific research plays a pivotal role in advancing society, stimulating economic growth, protecting the environment, and driving technological innovation. Understanding the key factors that lead to impactful research is thus crucial as it can steer the development of more effective policies to enhance the research enterprise. Extensive literature emphasises that the composition of a research team is vital for generating innovative and impactful scientific work. Many studies have focused on how team diversity, including aspects like ethnicity, gender, and international background, affects research outcomes. These types of diversities often correlate positively with the impact of research. In this paper, we investigate a less-explored dimension of diversity: the diversity of authors’ areas of expertise. This aspect has received limited attention, primarily due to the challenges involved in defining and measuring it. We present new AI-driven methods to quantify this diversity of expertise and conduct an extensive analysis of over 944,000 Computer Science papers. Specifically, this study investigates the relationship between the authors’ diversity of expertise and the number of citations their paper receives within the first five years. For each paper, we modelled the expertise of each individual author and then quantified the overall diversity within the author team. We then performed a statistical analysis that revealed a significant positive correlation between two diversity metrics and the number of citations received. This suggests that, in the field of Computer Science, diversity of expertise is a key driver of high-impact research.Source: SCIENTOMETRICS, vol. 131 (issue 2), pp. 1119-1146
DOI: 10.1007/s11192-026-05560-x
Metrics:
Salatino Angelo, Osborne Francesco, Recupero Reforgiato Diego, Angioni Simone, Motta Enrico
High-quality scientific research plays a pivotal role in advancing society, stimulating economic growth, protecting the environment, and driving technological innovation. Understanding the key factors that lead to impactful research is thus crucial as it can steer the development of more effective policies to enhance the research enterprise. Extensive literature emphasises that the composition of a research team is vital for generating innovative and impactful scientific work. Many studies have focused on how team diversity, including aspects like ethnicity, gender, and international background, affects research outcomes. These types of diversities often correlate positively with the impact of research. In this paper, we investigate a less-explored dimension of diversity: the diversity of authors’ areas of expertise. This aspect has received limited attention, primarily due to the challenges involved in defining and measuring it. We present new AI-driven methods to quantify this diversity of expertise and conduct an extensive analysis of over 944,000 Computer Science papers. Specifically, this study investigates the relationship between the authors’ diversity of expertise and the number of citations their paper receives within the first five years. For each paper, we modelled the expertise of each individual author and then quantified the overall diversity within the author team. We then performed a statistical analysis that revealed a significant positive correlation between two diversity metrics and the number of citations received. This suggests that, in the field of Computer Science, diversity of expertise is a key driver of high-impact research.Source: SCIENTOMETRICS, vol. 131 (issue 2), pp. 1119-1146
DOI: 10.1007/s11192-026-05560-x
Metrics:
See at:
CNR IRIS | link.springer.com | CNR IRIS
2026
Book
Open Access
Preface to 19th European Conference on Logics in Artificial Intelligence, JELIA 2025
Casini Giovanni, Dundua Besik, Kutsia Temur
These two volumes contain the proceedings of the 19th European Conference on Logics in Artificial Intelligence (JELIA 2025), held at Kutaisi International University, Kutaisi, Georgia from September 1 to 4, 2025.Source: LECTURE NOTES IN COMPUTER SCIENCE, vol. 16094
Casini Giovanni, Dundua Besik, Kutsia Temur
These two volumes contain the proceedings of the 19th European Conference on Logics in Artificial Intelligence (JELIA 2025), held at Kutaisi International University, Kutaisi, Georgia from September 1 to 4, 2025.Source: LECTURE NOTES IN COMPUTER SCIENCE, vol. 16094
See at:
CNR IRIS | link.springer.com | CNR IRIS
2026
Journal article
Open Access
Deploying conversational agents in virtual research environments: approaches and lessons learned
Assante Massimiliano, Candela Leonardo, Dell'Amico Andrea, Frosini Luca, Mangiacrapa Francesco, Oliviero Alfredo, Pagano Pasquale, Panichi Giancarlo, Peccerillo Biagio, Piccioli Tommaso, Procaccini Marco
Conversational agents have the potential to streamline tasks, provide support, and enhance user experience across various domains including Virtual Research Environments (VREs). The recent progress in conversational artificial intelligence and Large Language Models (LLMs) offers novel strategies for the development of these agents. This paper reports on the potential benefits, the challenges and the approaches resulting from concrete experiences in developing and equipping D4Science-based VREs with suitable conversational agents. The paper presents three successive implementation approaches and the resulting agent solution, each designed to address the limitations identified in the preceding iteration and to leverage the advantages offered by newer implementation and development options. The proposed approaches led to the progressive refinement of the agent design and functionality, resulting in DAVE, a conversational agent capable of securely interacting with multiple D4Science services and supporting a wide range of user workflows. The iterative process highlighted critical requirements—including authentication handling, usability, and extensibility—that can inform the design of conversational agents in similar research infrastructures. The study shows that conversational agents can effectively lower the barrier to accessing VRE functionalities and enhance user engagement. The resulting design principles and lessons learned provide a foundation for future work aimed at extending DAVE with an enhanced feedback mechanism and locally hosted LLM integration, and conducting systematic usability evaluations within active research communities.Source: SN COMPUTER SCIENCE, vol. 7
DOI: 10.1007/s42979-026-04863-3
Project(s): A federated European FAIR and Open Research Ecosystem for oceans, seas, coastal and inland waters, FOSSR—Fostering Open Science in Social Science Research
Metrics:
Assante Massimiliano, Candela Leonardo, Dell'Amico Andrea, Frosini Luca, Mangiacrapa Francesco, Oliviero Alfredo, Pagano Pasquale, Panichi Giancarlo, Peccerillo Biagio, Piccioli Tommaso, Procaccini Marco
Conversational agents have the potential to streamline tasks, provide support, and enhance user experience across various domains including Virtual Research Environments (VREs). The recent progress in conversational artificial intelligence and Large Language Models (LLMs) offers novel strategies for the development of these agents. This paper reports on the potential benefits, the challenges and the approaches resulting from concrete experiences in developing and equipping D4Science-based VREs with suitable conversational agents. The paper presents three successive implementation approaches and the resulting agent solution, each designed to address the limitations identified in the preceding iteration and to leverage the advantages offered by newer implementation and development options. The proposed approaches led to the progressive refinement of the agent design and functionality, resulting in DAVE, a conversational agent capable of securely interacting with multiple D4Science services and supporting a wide range of user workflows. The iterative process highlighted critical requirements—including authentication handling, usability, and extensibility—that can inform the design of conversational agents in similar research infrastructures. The study shows that conversational agents can effectively lower the barrier to accessing VRE functionalities and enhance user engagement. The resulting design principles and lessons learned provide a foundation for future work aimed at extending DAVE with an enhanced feedback mechanism and locally hosted LLM integration, and conducting systematic usability evaluations within active research communities.Source: SN COMPUTER SCIENCE, vol. 7
DOI: 10.1007/s42979-026-04863-3
Project(s): A federated European FAIR and Open Research Ecosystem for oceans, seas, coastal and inland waters, FOSSR—Fostering Open Science in Social Science Research
Metrics:
See at:
CNR IRIS | link.springer.com | CNR IRIS
2026
Journal article
Open Access
Editorial: data science and AI for marine science and the blue economy
Candela Leonardo, Pagano Pasquale, Bi Hongsheng, Schaap Dick
This editorial introduces the Special Collection “Data Science and AI for Marine Science and the Blue Economy” published in the International Journal of Data Science and Analytics. The collection explores how data-driven and AI-enabled approaches are advancing marine research, supporting operational monitoring, and enabling evidence-based decision-making across blue economy domains. The guest editors summarize the motivations for this initiative, briefly present the contributions included in the issue, and outline emerging themes and future perspectives in this evolving interdisciplinary field.Source: INTERNATIONAL JOURNAL OF DATA SCIENCE AND ANALYTICS, vol. 22 (issue 97)
DOI: 10.1007/s41060-026-01082-0
Project(s): Blue-Cloud 2026
Metrics:
Candela Leonardo, Pagano Pasquale, Bi Hongsheng, Schaap Dick
This editorial introduces the Special Collection “Data Science and AI for Marine Science and the Blue Economy” published in the International Journal of Data Science and Analytics. The collection explores how data-driven and AI-enabled approaches are advancing marine research, supporting operational monitoring, and enabling evidence-based decision-making across blue economy domains. The guest editors summarize the motivations for this initiative, briefly present the contributions included in the issue, and outline emerging themes and future perspectives in this evolving interdisciplinary field.Source: INTERNATIONAL JOURNAL OF DATA SCIENCE AND ANALYTICS, vol. 22 (issue 97)
DOI: 10.1007/s41060-026-01082-0
Project(s): Blue-Cloud 2026
Metrics:
See at:
CNR IRIS | link.springer.com | CNR IRIS
2026
Conference article
Open Access
Enhancing interoperability of SPARQL endpoints: RESTful and OAI-PMH API for the DH-ATLAS project
Rubin Giorgia, Bardi Alessia, Del Gratta Riccardo, Del Grosso Angelo Mario
Digital repositories leveraging RDF and Linked Data for Cultural Heritage metadata face a critical challenge: their native SPARQL endpoints often create silos because major aggregators rely on the OAI-PMH protocol for harvesting, and developers prefer RESTful APIs. This divergence undermines the visibility and FAIRness (Findable, Accessible, Interoperable, Reusable) of scholarly resources. This paper details the unified access strategy implemented for the DH-ATLAS project, which generated an Ontology and Knowledge Graph focused on Italian Digital Cultural Heritage research. To ensure broad dissemination and resource reuse without duplicating data, DH-ATLAS developed two configurable and modular software components to implement compatibility with the OpenAIRE guidelines and REST clients. Together, these components establish a cohesive and reusable solution for integrating semantic repositories with diverse metadata consumers, promoting the long-term sustainability and broader accessibility of the DH-ATLAS Knowledge Graph and serving as a model for other RDF infrastructures.Project(s): The ATLAS of Italian Digital Humanities: a dynamic knowledge graph of digital scholarly research on Italian Cultural Heritage
Rubin Giorgia, Bardi Alessia, Del Gratta Riccardo, Del Grosso Angelo Mario
Digital repositories leveraging RDF and Linked Data for Cultural Heritage metadata face a critical challenge: their native SPARQL endpoints often create silos because major aggregators rely on the OAI-PMH protocol for harvesting, and developers prefer RESTful APIs. This divergence undermines the visibility and FAIRness (Findable, Accessible, Interoperable, Reusable) of scholarly resources. This paper details the unified access strategy implemented for the DH-ATLAS project, which generated an Ontology and Knowledge Graph focused on Italian Digital Cultural Heritage research. To ensure broad dissemination and resource reuse without duplicating data, DH-ATLAS developed two configurable and modular software components to implement compatibility with the OpenAIRE guidelines and REST clients. Together, these components establish a cohesive and reusable solution for integrating semantic repositories with diverse metadata consumers, promoting the long-term sustainability and broader accessibility of the DH-ATLAS Knowledge Graph and serving as a model for other RDF infrastructures.Project(s): The ATLAS of Italian Digital Humanities: a dynamic knowledge graph of digital scholarly research on Italian Cultural Heritage
See at:
CNR IRIS | ircdl2026.unimore.it | CNR IRIS
2026
Contribution to conference
Open Access
A preliminary analysis of high water events in Venice based on multi-decadal observations and clustering
Cardillo Franco Alberto, Andrigo Angela, De Biasio Francesco, Debole Franca, Favaro Marco, Papa Alvise, Straccia Umberto, Vignudelli Stefano
High water events in Venice are a recurrent phenomenon, as the city is located only slightly above mean sea level and is directly in"uenced by water-level variations within the lagoon. Repeated "ooding has signi!cant economic and social impacts, limits pedestrian and naval tra#c and contributes to the degradation of buildings and cultural heritage. Current forecasting systems primarily estimate water levels and peak values, and these are typically estimated at a limited number of locations. Data-driven approaches, in particular Machine Learning (ML) methods, analyze historical data without relying on prede!ned, human-designed model structures. We present a preliminary analysis based on several clustering approaches, including k-means, DBSCAN, and deep learning–based methods, applied to a multi-decadal atmospheric dataset and to the longest available reconstructed hourly sea-level records for the northern Adriatic Sea, specifically developed for this study.Project(s): Collaborazione scientifica ILC - CPSM - ISTI
Cardillo Franco Alberto, Andrigo Angela, De Biasio Francesco, Debole Franca, Favaro Marco, Papa Alvise, Straccia Umberto, Vignudelli Stefano
High water events in Venice are a recurrent phenomenon, as the city is located only slightly above mean sea level and is directly in"uenced by water-level variations within the lagoon. Repeated "ooding has signi!cant economic and social impacts, limits pedestrian and naval tra#c and contributes to the degradation of buildings and cultural heritage. Current forecasting systems primarily estimate water levels and peak values, and these are typically estimated at a limited number of locations. Data-driven approaches, in particular Machine Learning (ML) methods, analyze historical data without relying on prede!ned, human-designed model structures. We present a preliminary analysis based on several clustering approaches, including k-means, DBSCAN, and deep learning–based methods, applied to a multi-decadal atmospheric dataset and to the longest available reconstructed hourly sea-level records for the northern Adriatic Sea, specifically developed for this study.Project(s): Collaborazione scientifica ILC - CPSM - ISTI
2026
Journal article
Open Access
Supporting open science in virtual research environments: the DAVE experience
Dell’amico Andrea, Oliviero Alfredo, Panichi Giancarlo, Peccerillo Biagio, Procaccini Marco
DAVE, a conversational assistant integrated into D4Science Virtual Research Environments, simplifies access to services and supports Open Science workflows through natural language interaction. Virtual Research Environments (VREs) provide integrated access to data, services, and collaboration tools for open and data-intensive research. Building on the D4Science Virtual Research Environments, which support over 230 VREs and around 28,000 users worldwide, our work on conversational agents has led to the development of DAVE (D4Science Assistant for Virtual Research Environments), a system designed to assist researchers directly within their workflows.Source: ERCIM NEWS, vol. 144, pp. 39-40
Dell’amico Andrea, Oliviero Alfredo, Panichi Giancarlo, Peccerillo Biagio, Procaccini Marco
DAVE, a conversational assistant integrated into D4Science Virtual Research Environments, simplifies access to services and supports Open Science workflows through natural language interaction. Virtual Research Environments (VREs) provide integrated access to data, services, and collaboration tools for open and data-intensive research. Building on the D4Science Virtual Research Environments, which support over 230 VREs and around 28,000 users worldwide, our work on conversational agents has led to the development of DAVE (D4Science Assistant for Virtual Research Environments), a system designed to assist researchers directly within their workflows.Source: ERCIM NEWS, vol. 144, pp. 39-40
See at:
ercim-news.ercim.eu | CNR IRIS | CNR IRIS
2026
Other
Open Access
Il trattamento dei dati personali nella ricerca: evento di formazione H2IOSC ai giovani ricercatori del settore SSH
Luzietti Roberta Bianca, Ottaviani Roberta
Rapporto Tecnico relativo agli output emergenti da una formazione a giovani ricercatori erogata nell'ambito del Progetto H2IOSC (PNRR IR)Project(s): Humanities and cultural Heritage Italian Open Science Cloud funded by the European Union
Luzietti Roberta Bianca, Ottaviani Roberta
Rapporto Tecnico relativo agli output emergenti da una formazione a giovani ricercatori erogata nell'ambito del Progetto H2IOSC (PNRR IR)Project(s): Humanities and cultural Heritage Italian Open Science Cloud funded by the European Union
2025
Journal article
Open Access
A survey of knowledge organization systems of research fields: resources and challenges
Salatino A., Aggarwal T., Mannocci A., Osborne F., Motta E.
Knowledge organization systems (KOSs), such as term lists, thesauri, taxonomies, andontologies, play a fundamental role in categorizing, managing, and retrieving information. Inthe academic domain, KOSs are often adopted for representing research areas and theirrelationships, primarily aiming to classify research articles, academic courses, patents, books,scientific venues, domain experts, grants, software, experiment materials, and several otherrelevant products and agents. These structured representations of research areas, widelyembraced by many academic fields, have proven effective in empowering AI-based systems toenhance the retrievability of relevant documents, enable advanced analytic solutions toquantify the impact of academic research, and analyze and forecast research dynamics.We aim to present a comprehensive survey of the current KOS for academic disciplines.We analyzed and compared 45 KOSs according to five main dimensions: scope, structure,curation, usage, and links to other KOSs. Our results reveal a very heterogeneous scenario interms of scope, scale, quality, and usage, highlighting the need for more integrated solutionsfor representing research knowledge across academic fields. We conclude by discussing themain challenges and the most promising future directions.Source: QUANTITATIVE SCIENCE STUDIES, vol. 6, pp. 567-610
DOI: 10.1162/qss_a_00363
Metrics:
Salatino A., Aggarwal T., Mannocci A., Osborne F., Motta E.
Knowledge organization systems (KOSs), such as term lists, thesauri, taxonomies, andontologies, play a fundamental role in categorizing, managing, and retrieving information. Inthe academic domain, KOSs are often adopted for representing research areas and theirrelationships, primarily aiming to classify research articles, academic courses, patents, books,scientific venues, domain experts, grants, software, experiment materials, and several otherrelevant products and agents. These structured representations of research areas, widelyembraced by many academic fields, have proven effective in empowering AI-based systems toenhance the retrievability of relevant documents, enable advanced analytic solutions toquantify the impact of academic research, and analyze and forecast research dynamics.We aim to present a comprehensive survey of the current KOS for academic disciplines.We analyzed and compared 45 KOSs according to five main dimensions: scope, structure,curation, usage, and links to other KOSs. Our results reveal a very heterogeneous scenario interms of scope, scale, quality, and usage, highlighting the need for more integrated solutionsfor representing research knowledge across academic fields. We conclude by discussing themain challenges and the most promising future directions.Source: QUANTITATIVE SCIENCE STUDIES, vol. 6, pp. 567-610
DOI: 10.1162/qss_a_00363
Metrics:
See at:
direct.mit.edu | CNR IRIS | CNR IRIS
2025
Conference article
Open Access
ATLAS: a data model for describing FAIR Digital Humanities research outcomes
Martignano C., Rubin G., Giacomini S., Bardi A., Buzzoni M., Daquino M., Del Gratta R., Del Grosso A. M., Fischer F., Rosselli Del Turco R., Tomasi F.
This paper addresses the challenges of cataloguing and representing Digital Humanities (DH) research outputs within the framework of FAIR principles. Despite advancements in Semantic Web technologies and data aggregators, the scholarly community still lacks unified frameworks and domain-specific models to describe heterogeneous outputs such as digital editions, textual collections, and other scholarly resources published as Linked Open Data. The ATLAS project proposes an ontology and a knowledge graph to bridge these gaps. The methodology includes metadata modeling based on existing frameworks and novel extensions, supported by pilot studies on the Italian cultural heritage. The project, currently in its initial version, aims to enhance metadata interoperability and data accessibility, contributing to the optimization of cataloguing practices and the development of guidelines for the discovery and reuse of DH resources while providing a model applicable beyond the Italian context.Source: QUADERNI DI UMANISTICA DIGITALE, pp. 440-447. Verona, Italy, 11-13/06/2025
DOI: 10.6092/unibo/amsacta/8380
Project(s): The ATLAS of Italian Digital Humanities: a dynamic knowledge graph of digital scholarly research on Italian Cultural Heritage
Metrics:
Martignano C., Rubin G., Giacomini S., Bardi A., Buzzoni M., Daquino M., Del Gratta R., Del Grosso A. M., Fischer F., Rosselli Del Turco R., Tomasi F.
This paper addresses the challenges of cataloguing and representing Digital Humanities (DH) research outputs within the framework of FAIR principles. Despite advancements in Semantic Web technologies and data aggregators, the scholarly community still lacks unified frameworks and domain-specific models to describe heterogeneous outputs such as digital editions, textual collections, and other scholarly resources published as Linked Open Data. The ATLAS project proposes an ontology and a knowledge graph to bridge these gaps. The methodology includes metadata modeling based on existing frameworks and novel extensions, supported by pilot studies on the Italian cultural heritage. The project, currently in its initial version, aims to enhance metadata interoperability and data accessibility, contributing to the optimization of cataloguing practices and the development of guidelines for the discovery and reuse of DH resources while providing a model applicable beyond the Italian context.Source: QUADERNI DI UMANISTICA DIGITALE, pp. 440-447. Verona, Italy, 11-13/06/2025
DOI: 10.6092/unibo/amsacta/8380
Project(s): The ATLAS of Italian Digital Humanities: a dynamic knowledge graph of digital scholarly research on Italian Cultural Heritage
Metrics:
See at:
amsacta.unibo.it | CNR IRIS | AMS Acta | CNR IRIS