Fabio Vitali
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Fabio Vitali has a total of 164 co-authors. Here are the top co-authors by number of shared publications:
- Silvio Peroni
University of Bologna
39 shared publications - Angelo Di Iorio
Department of Computer Science and Engineering, University of Bologna, Bologna, Italy
37 shared publications - Paolo Ciancarini
University of Bologna
16 shared publications - Francesca Tomasi
Alma Mater Studiorum University of Bologna
12 shared publications - Monica Palmirani
University of Bologna
9 shared publications - Francesco Sovrano
Università Della Svizzera italiana, Lugano, Switzerland
8 shared publications
Latest Publications
- Open Access
- 5 October 2026
- Published by Association for Computing Machinery (ACM)
- p. 370-374
Enabling AI to Find Multimodal Signals of Storytelling Cross-Culturally: A Multi-Step Approach
- 5 October 2026
- Published by Association for Computing Machinery (ACM)
- p. 370-374
Open AccessEnabling AI to Find Multimodal Signals of Storytelling Cross-Culturally: A Multi-Step Approach
- Open Access
- 18 August 2026
- Published by Association for Computing Machinery (ACM)
- p. 1-4
Twenty Years of Akoma Ntoso: Alive and Kicking!
- 18 August 2026
- Published by Association for Computing Machinery (ACM)
- p. 1-4
Open AccessTwenty Years of Akoma Ntoso: Alive and Kicking!
- Open Access
Knowledge Graphs Generation from Cultural Heritage Texts: Combining LLMs and Ontological Engineering for Scholarly Debates
Cultural Heritage texts contain rich knowledge that is difficult to query systematically due to the challenges of converting unstructured discourse into structured Knowledge Graphs (KGs). This paper introduces ATR4CH (Adaptive Text-to-RDF for Cultural Heritage), a systematic five-step methodology for Large Language Model-based Knowledge Extraction from Cultural Heritage documents. We validate the methodology through a case study on authenticity assessment debates. Methodology - ATR4CH combines annotation models, ontological frameworks, and LLM-based extraction through iterative development: foundational analysis, annotation schema development, pipeline architecture, integration refinement, and comprehensive evaluation. We demonstrate the approach using Wikipedia articles about disputed items (documents, artifacts...), implementing a sequential pipeline with three LLMs (Claude Sonnet 3.7, Llama 3.3 70B, GPT-4o-mini). Findings - The methodology successfully extracts complex Cultural Heritage knowledge: 0.96-0.99 F1 for metadata extraction, 0.7-0.8 F1 for entity recognition, 0.65-0.75 F1 for hypothesis extraction, 0.95-0.97 for evidence extraction, and 0.62 G-EVAL for discourse representation. Smaller models performed competitively, enabling cost-effective deployment. Originality - This is the first systematic methodology for coordinating LLM-based extraction with Cultural Heritage ontologies. ATR4CH provides a replicable framework adaptable across CH domains and institutional resources. Research Limitations - The produced KG is limited to Wikipedia articles. While the results are encouraging, human oversight is necessary during post-processing. Practical Implications - ATR4CH enables Cultural Heritage institutions to systematically convert textual knowledge into queryable KGs, supporting automated metadata enrichment and knowledge discovery.
Open AccessKnowledge Graphs Generation from Cultural Heritage Texts: Combining LLMs and Ontological Engineering for Scholarly Debates
Cultural Heritage texts contain rich knowledge that is difficult to query systematically due to the challenges of converting unstructured discourse into structured Knowledge Graphs (KGs). This paper introduces ATR4CH (Adaptive Text-to-RDF for Cultural Heritage), a systematic five-step methodology for Large Language Model-based Knowledge Extraction from Cultural Heritage documents. We validate the methodology through a case study on authenticity assessment debates. Methodology - ATR4CH combines annotation models, ontological frameworks, and LLM-based extraction through iterative development: foundational analysis, annotation schema development, pipeline architecture, integration refinement, and comprehensive evaluation. We demonstrate the approach using Wikipedia articles about disputed items (documents, artifacts...), implementing a sequential pipeline with three LLMs (Claude Sonnet 3.7, Llama 3.3 70B, GPT-4o-mini). Findings - The methodology successfully extracts complex Cultural Heritage knowledge: 0.96-0.99 F1 for metadata extraction, 0.7-0.8 F1 for entity recognition, 0.65-0.75 F1 for hypothesis extraction, 0.95-0.97 for evidence extraction, and 0.62 G-EVAL for discourse representation. Smaller models performed competitively, enabling cost-effective deployment. Originality - This is the first systematic methodology for coordinating LLM-based extraction with Cultural Heritage ontologies. ATR4CH provides a replicable framework adaptable across CH domains and institutional resources. Research Limitations - The produced KG is limited to Wikipedia articles. While the results are encouraging, human oversight is necessary during post-processing. Practical Implications - ATR4CH enables Cultural Heritage institutions to systematically convert textual knowledge into queryable KGs, supporting automated metadata enrichment and knowledge discovery.
- Open Access
- 21 June 2025
- Published by Association for Computing Machinery (ACM) in Journal on Computing and Cultural Heritage
- Vol. 18 (2) , 1-22
The Semantic Digital Edition of Aldo Moro’s Writings: A Workflow Supporting Data Sharing and Replicability
Digital editions have been long recognized as significant scholarly outputs, reflecting a tradition dating back to computational philology and evolving to encompass comprehensive literary and scientific knowledge management on the web. However, debates still persist around how to use data models, technical tools, and interface design to implement and document such resources so as to make them as replicable and usable as possible. Given these premises, this article focuses on presenting a semantic digital edition workflow, and, specifically, examines three crucial aspects: the integration of Semantic Web technologies for enriching textual representation; the development of a web-based markup tool that enables domain experts to produce crowdsourced markup; and the production of comprehensive technical documentation to ensure replicability. To do so, the article details how the workflow was applied to create a digital edition of Aldo Moro’s works within a project funded by the Italian Ministry for Cultural Heritage and Activities and for Tourism. The article outlines the workflow stages, including survey and transcription, data modeling, markup and metadata insertion, web design, and documentation, underscoring the importance of formalized methods in developing digital scholarly editions. By adhering to established editorial and technological standards, the edition offers a nuanced exploration of Moro’s writings, leveraging semantic technologies to provide a platform for scholarly engagement and exploration on the web.
- 21 June 2025
- Published by Association for Computing Machinery (ACM) in Journal on Computing and Cultural Heritage
- Vol. 18 (2) , 1-22
Open AccessThe Semantic Digital Edition of Aldo Moro’s Writings: A Workflow Supporting Data Sharing and Replicability
Digital editions have been long recognized as significant scholarly outputs, reflecting a tradition dating back to computational philology and evolving to encompass comprehensive literary and scientific knowledge management on the web. However, debates still persist around how to use data models, technical tools, and interface design to implement and document such resources so as to make them as replicable and usable as possible. Given these premises, this article focuses on presenting a semantic digital edition workflow, and, specifically, examines three crucial aspects: the integration of Semantic Web technologies for enriching textual representation; the development of a web-based markup tool that enables domain experts to produce crowdsourced markup; and the production of comprehensive technical documentation to ensure replicability. To do so, the article details how the workflow was applied to create a digital edition of Aldo Moro’s works within a project funded by the Italian Ministry for Cultural Heritage and Activities and for Tourism. The article outlines the workflow stages, including survey and transcription, data modeling, markup and metadata insertion, web design, and documentation, underscoring the importance of formalized methods in developing digital scholarly editions. By adhering to established editorial and technological standards, the edition offers a nuanced exploration of Moro’s writings, leveraging semantic technologies to provide a platform for scholarly engagement and exploration on the web.
- Open Access
- 22 May 2025
- Published by Edizioni Ca Foscari
The Representation of Historical Uncertainties as the Outcome of Competing and Incompatible Certainties
This study delves into the ontological implications associated with ‘data’ and ‘capta’, highlighting the profound importance of contextual factors and interpretation within the realm of critical discourse. More specifically, ‘data’, encapsulating the notion of objective observations of facts, differ from ‘capta’, which encompass selections, opinions, controversies, and debates. These capta elements contribute interest and added value to scholarly knowledge. The formalised separation of data and capta proves to be pivotal in shaping research practices conducive to a critical approach to knowledge production. Upon an evaluation of the efficacy of existing technologies in addressing this issue, the Conjectures model is introduced as a potential solution. This model aims to explicitly represent capta while concurrently facilitating the evolution of critical discourse within a one Knowledge Graph designed to capture complex information within a particular domain. To underscore our point, an illustrative example is drawn from the field of historiography.
- 22 May 2025
- Published by Edizioni Ca Foscari
Open AccessThe Representation of Historical Uncertainties as the Outcome of Competing and Incompatible Certainties
This study delves into the ontological implications associated with ‘data’ and ‘capta’, highlighting the profound importance of contextual factors and interpretation within the realm of critical discourse. More specifically, ‘data’, encapsulating the notion of objective observations of facts, differ from ‘capta’, which encompass selections, opinions, controversies, and debates. These capta elements contribute interest and added value to scholarly knowledge. The formalised separation of data and capta proves to be pivotal in shaping research practices conducive to a critical approach to knowledge production. Upon an evaluation of the efficacy of existing technologies in addressing this issue, the Conjectures model is introduced as a potential solution. This model aims to explicitly represent capta while concurrently facilitating the evolution of critical discourse within a one Knowledge Graph designed to capture complex information within a particular domain. To underscore our point, an illustrative example is drawn from the field of historiography.
- Open Access
- 19 April 2025
- Published by Association for Computing Machinery (ACM) in Journal on Computing and Cultural Heritage
- Vol. 18 (2) , 1-20
A Methodology for Enhancing the Accessibility of Scientific Museum’s Collections (SCAMP)
Museums are full of beautiful objects with high historical value. Sometimes such objects are static in their nature (e.g., statues), but often they are interactive (e.g., tools or manipulable artifacts). Very often they are accompanied by the classic sign “do not touch, just watch” or are enclosed in glass cases, hindering the full experience of their intended use. Due to high restoration and preservation costs, some of these artifacts are also relegated to storage, limiting access to the public. This poses a dual accessibility issue: the general public cannot interact with these objects as originally intended, and blind or visually impaired individuals must rely solely on textual descriptions or audio guides. Additionally, certain artifacts still hold significant societal value if they could be interactively experienced again. To address this, we propose the Scientific Collections Accessibility Making Process (SCAMP) methodology. By leveraging 3D printing and digital technologies, SCAMP is a robust, generic framework to “make accessible again” interactive museum objects. It creates accurate replicas and a multimodal experience centered around the recreated artifacts. Specific guidelines ensure that the replica and the interactive user experience maintain fidelity to the original objects, enhancing (rather than altering) their characteristics. Unlike many single-case discussions in the literature, SCAMP focuses on a scalable, theoretical basis that can be applied to similar artifacts. Beyond accurately reproducing shapes and colors, SCAMP also emphasizes tactile fidelity, conveying sensations close to those provided by the original item. By applying the SCAMP methodology, interactive objects of high historical value can be made accessible to everyone—both the general public and individuals who are blind or visually impaired. Accessibility in SCAMP is essentially about ensuring an inclusive user experience for the interactive content, for which specific design rules are provided. We demonstrate SCAMP through two Brendel Models from the University of Bologna, concluding with user testing. While there is room for improvement, results show that SCAMP successfully meets its accessibility and fidelity goals, with users expressing enthusiasm for the final outcome.
- 19 April 2025
- Published by Association for Computing Machinery (ACM) in Journal on Computing and Cultural Heritage
- Vol. 18 (2) , 1-20
Open AccessA Methodology for Enhancing the Accessibility of Scientific Museum’s Collections (SCAMP)
Museums are full of beautiful objects with high historical value. Sometimes such objects are static in their nature (e.g., statues), but often they are interactive (e.g., tools or manipulable artifacts). Very often they are accompanied by the classic sign “do not touch, just watch” or are enclosed in glass cases, hindering the full experience of their intended use. Due to high restoration and preservation costs, some of these artifacts are also relegated to storage, limiting access to the public. This poses a dual accessibility issue: the general public cannot interact with these objects as originally intended, and blind or visually impaired individuals must rely solely on textual descriptions or audio guides. Additionally, certain artifacts still hold significant societal value if they could be interactively experienced again. To address this, we propose the Scientific Collections Accessibility Making Process (SCAMP) methodology. By leveraging 3D printing and digital technologies, SCAMP is a robust, generic framework to “make accessible again” interactive museum objects. It creates accurate replicas and a multimodal experience centered around the recreated artifacts. Specific guidelines ensure that the replica and the interactive user experience maintain fidelity to the original objects, enhancing (rather than altering) their characteristics. Unlike many single-case discussions in the literature, SCAMP focuses on a scalable, theoretical basis that can be applied to similar artifacts. Beyond accurately reproducing shapes and colors, SCAMP also emphasizes tactile fidelity, conveying sensations close to those provided by the original item. By applying the SCAMP methodology, interactive objects of high historical value can be made accessible to everyone—both the general public and individuals who are blind or visually impaired. Accessibility in SCAMP is essentially about ensuring an inclusive user experience for the interactive content, for which specific design rules are provided. We demonstrate SCAMP through two Brendel Models from the University of Bologna, concluding with user testing. While there is room for improvement, results show that SCAMP successfully meets its accessibility and fidelity goals, with users expressing enthusiasm for the final outcome.
- Open Access
KwicKwocKwac, a tool for rapidly generating concordances and marking up a literary text
This paper introduces KwicKwocKwac 1.0 (KwicKK), a web application designed to enhance the annotation and enrichment of digital texts in the humanities. KwicKK provides a user-friendly interface that enables scholars and researchers to perform semi-automatic markup of textual documents, facilitating the identification of relevant entities such as people, organizations, and locations. Key functionalities include the visualization of annotated texts using KeyWord in Context (KWIC), KeyWord Out Of Context (KWOC), and KeyWord After Context (KWAC) methodologies, alongside automatic disambiguation of generic references and integration with Wikidata for Linked Open Data connections. The application supports metadata input and offers multiple download formats, promoting accessibility and ease of use. Developed primarily for the National Edition of Aldo Moro's works, KwicKK aims to lower the technical barriers for users while fostering deeper engagement with digital scholarly resources. The architecture leverages contemporary web technologies, ensuring scalability and reliability. Future developments will explore user experience enhancements, collaborative features, and integration of additional data sources.
Open AccessKwicKwocKwac, a tool for rapidly generating concordances and marking up a literary text
This paper introduces KwicKwocKwac 1.0 (KwicKK), a web application designed to enhance the annotation and enrichment of digital texts in the humanities. KwicKK provides a user-friendly interface that enables scholars and researchers to perform semi-automatic markup of textual documents, facilitating the identification of relevant entities such as people, organizations, and locations. Key functionalities include the visualization of annotated texts using KeyWord in Context (KWIC), KeyWord Out Of Context (KWOC), and KeyWord After Context (KWAC) methodologies, alongside automatic disambiguation of generic references and integration with Wikidata for Linked Open Data connections. The application supports metadata input and offers multiple download formats, promoting accessibility and ease of use. Developed primarily for the National Edition of Aldo Moro's works, KwicKK aims to lower the technical barriers for users while fostering deeper engagement with digital scholarly resources. The architecture leverages contemporary web technologies, ensuring scalability and reliability. Future developments will explore user experience enhancements, collaborative features, and integration of additional data sources.
- Open Access
- 11 September 2024
- Published by IOS Press
- p. 370-385
What Do We Annotate When We Annotate? Towards a Multi-Level Approach to Semantic Annotations
Purpose: Annotating is considered a ‘scholarly primitive’ among different fields in the humanities. Nevertheless, the debate on digital annotations has mostly focused on the annotation of textual data, whereas existing models for representing annotations of images still lack sufficient semantic richness to anchor the annotation itself to multiple conceptual levels. We address the challenge of defining a data model to overcome the problem of ‘semantic deficit’ in this application domain. Finally, we implement an annotation client for testing multi-level semantic annotations. Methodology: To define a data model for representing digital annotations, we analysed applications which support annotation images through IIIF protocol, focusing on digital representations of palimpsests. We then extended the Web Annotation Data Model by introducing domain standards such as LRMer, CIDOC-CRM, and HiCO. We also validated the model through SPARQL queries corresponding to five competency questions to report on satisfiability. Finally, we developed a prototype annotation client as a plugin for Mirador to evaluate its performances in real-world scenarios. Findings: The results indicate that our model can effectively disambiguate between a target image and multiple conceptual levels of the entity itself, proving to be decisive in the representation of entities that coexist in the same material item (e.g., palimpsests). Additionally, the model allows users to describe annotations as interpretative acts, incorporating scholarly criteria and multiple viewpoints. An interface plugin enables scholars without technical expertise to create structured annotations that comply with the model. Value: The proposed approach facilitates the detailed management of the relationships between digital resources and their annotations, improving interoperability and information accessibility in the Semantic Web domain. Future developments will concern further extensions of the model, considering information about versioning, provenance, and authoritativeness of the digital annotations on images, as well as support for meta-annotations and iconological levels of interpretation.
- 11 September 2024
- Published by IOS Press
- p. 370-385
Open AccessWhat Do We Annotate When We Annotate? Towards a Multi-Level Approach to Semantic Annotations
Purpose: Annotating is considered a ‘scholarly primitive’ among different fields in the humanities. Nevertheless, the debate on digital annotations has mostly focused on the annotation of textual data, whereas existing models for representing annotations of images still lack sufficient semantic richness to anchor the annotation itself to multiple conceptual levels. We address the challenge of defining a data model to overcome the problem of ‘semantic deficit’ in this application domain. Finally, we implement an annotation client for testing multi-level semantic annotations. Methodology: To define a data model for representing digital annotations, we analysed applications which support annotation images through IIIF protocol, focusing on digital representations of palimpsests. We then extended the Web Annotation Data Model by introducing domain standards such as LRMer, CIDOC-CRM, and HiCO. We also validated the model through SPARQL queries corresponding to five competency questions to report on satisfiability. Finally, we developed a prototype annotation client as a plugin for Mirador to evaluate its performances in real-world scenarios. Findings: The results indicate that our model can effectively disambiguate between a target image and multiple conceptual levels of the entity itself, proving to be decisive in the representation of entities that coexist in the same material item (e.g., palimpsests). Additionally, the model allows users to describe annotations as interpretative acts, incorporating scholarly criteria and multiple viewpoints. An interface plugin enables scholars without technical expertise to create structured annotations that comply with the model. Value: The proposed approach facilitates the detailed management of the relationships between digital resources and their annotations, improving interoperability and information accessibility in the Semantic Web domain. Future developments will concern further extensions of the model, considering information about versioning, provenance, and authoritativeness of the digital annotations on images, as well as support for meta-annotations and iconological levels of interpretation.
- Open Access
- 4 September 2024
- Published by Association for Computing Machinery (ACM)
- p. 1
AAA: A blockchain-based architecture for ethical, robust authenticated anonymity
In the past years, online anonymity has attracted strong criticism for its role in shielding online crimes such as cyberbullying, fake news, money laundering, and pedo-pornography. Yet, it also has historically strong supporters who emphasize the necessity of a safe haven for carrying out legal and ethical activities that should not be associated with our real-life personas. We define authenticated anonymity as the possibility of using anonymous accounts that cannot be associated with the real identity of their owner unless a criminal act is being performed through them. Blockchain technology represents a good means for managing this complexity in a secure and trustworthy manner. Several solutions exist in the literature and on the market for anonymous identity, but they confer too much power to their owners, who can decide what to reveal about themselves in total autonomy (self-sovereign identities). In this paper, we present the Authenticated Anonymity Architecture (AAA), a blockchain-based solution for creating authenticated anonymous identities, where the mappings between official and anonymous identities can only be revealed after the necessary consensus of multiple different actors on the blockchain, evaluating the appropriateness and ethicality of the request. We mathematically modeled the architecture and conducted some analytical evaluations, showing that our proposal is resilient and fault-tolerant, even in the case of a huge number of identities managed.
- 4 September 2024
- Published by Association for Computing Machinery (ACM)
- p. 1
Open AccessAAA: A blockchain-based architecture for ethical, robust authenticated anonymity
In the past years, online anonymity has attracted strong criticism for its role in shielding online crimes such as cyberbullying, fake news, money laundering, and pedo-pornography. Yet, it also has historically strong supporters who emphasize the necessity of a safe haven for carrying out legal and ethical activities that should not be associated with our real-life personas. We define authenticated anonymity as the possibility of using anonymous accounts that cannot be associated with the real identity of their owner unless a criminal act is being performed through them. Blockchain technology represents a good means for managing this complexity in a secure and trustworthy manner. Several solutions exist in the literature and on the market for anonymous identity, but they confer too much power to their owners, who can decide what to reveal about themselves in total autonomy (self-sovereign identities). In this paper, we present the Authenticated Anonymity Architecture (AAA), a blockchain-based solution for creating authenticated anonymous identities, where the mappings between official and anonymous identities can only be revealed after the necessary consensus of multiple different actors on the blockchain, evaluating the appropriateness and ethicality of the request. We mathematically modeled the architecture and conducted some analytical evaluations, showing that our proposal is resilient and fault-tolerant, even in the case of a huge number of identities managed.
- Open Access
Structuring Authenticity Assessments on Historical Documents using LLMs
Given the wide use of forgery throughout history, scholars have and are continuously engaged in assessing the authenticity of historical documents. However, online catalogues merely offer descriptive metadata for these documents, relegating discussions about their authenticity to free-text formats, making it difficult to study these assessments at scale. This study explores the generation of structured data about documents' authenticity assessment from natural language texts. Our pipeline exploits Large Language Models (LLMs) to select, extract and classify relevant claims about the topic without the need for training, and Semantic Web technologies to structure and type-validate the LLM's results. The final output is a catalogue of documents whose authenticity has been debated, along with scholars' opinions on their authenticity. This process can serve as a valuable resource for integration into catalogues, allowing room for more intricate queries and analyses on the evolution of these debates over centuries.
Open AccessStructuring Authenticity Assessments on Historical Documents using LLMs
Given the wide use of forgery throughout history, scholars have and are continuously engaged in assessing the authenticity of historical documents. However, online catalogues merely offer descriptive metadata for these documents, relegating discussions about their authenticity to free-text formats, making it difficult to study these assessments at scale. This study explores the generation of structured data about documents' authenticity assessment from natural language texts. Our pipeline exploits Large Language Models (LLMs) to select, extract and classify relevant claims about the topic without the need for training, and Semantic Web technologies to structure and type-validate the LLM's results. The final output is a catalogue of documents whose authenticity has been debated, along with scholars' opinions on their authenticity. This process can serve as a valuable resource for integration into catalogues, allowing room for more intricate queries and analyses on the evolution of these debates over centuries.