Modern distributed systems are increasingly complex and interconnected. Delivered on the cloud-edge continuum, boosted by 5G, built on AI, and deployed in critical domains (health, transportation…). These systems produce, store, share, and analyze huge amount of data, posing strong requirements on their quality and privacy, and requesting new governance solutions.

AI, foundational models, LNP add complexity on top of these systems, while being an opportunity to connect heterogeneous systems. These systems also require new verification techniques proving their (non-functional) behavior.

In this context, data governance, including privacy, data protection, data quality, and reliability , become fundamental requirements all backed by the need of interoperability.

The role of data: open data vs free-to-use data, regulations (eu vs non-EU)

Presenters

Ernesto Damiani, President, CINI, Full Professor, Università degli Studi di Milano, Italy, Director, Center for Cyber-Physical System, Khalifa University, UAE, CEO MoonCloud srl
Bio: President of the National Interuniversity Consortium for Computer Science. EIC of the International Journal on Big Data and of the International Journal of Knowledge and Learning. Associate editor of the IEEE Transactions on Service-oriented Computing and of the IEEE Transactions on Fuzzy Systems. ACM Distinguished Scientist and recipient of the Chester Sall Award from the IEEE Industrial Electronics Society. He received the doctorate honoris causa from INSA - Lyon (France) for his contribution to Big Data teaching and research. His research interests include Artificial Intelligence, Machine Learning, Cyber-physical systems, secure service-oriented architectures, privacy-preserving big data analytics and cyber-physical systems security, where he published more than 700 scientific papers and books.

Valter Fraccaro, Presidente Fondazione SAIHub Siena,
Bio: He is President of the Fondazione SAIHUB, a foundation focused on the application of artificial intelligence in the life sciences sector (agri-food, pharmaceuticals, biotechnology, medical devices). His career began in 1987 with a focus on digital cataloguing of cultural heritage. In the IT sector, he has worked as a manager in Italian and international companies and at the University of Padua as Director of the Centre for Information Systems. In the past, he founded two companies: the first dedicated to digital innovation projects in industry, the second specialised in the IT management of scientific data (founded in 2015, the first in Europe dedicated to this topic). Since 2017, he has started his collaboration with the University of Siena and then with pharmaceutical and technology companies, dealing with artificial intelligence projects and their funding. As a professional or as chairman of the Fondazione SAIHUB, he has collaborated in the design of technology projects that have channelled around 100 million euros of European funds to public and private initiatives in Italy. He is a consultant and trainer for private companies, foundations, consultancies, cultural institutions and universities on issues of planning, ethics, design and sustainability of innovation, with a focus on Artificial Intelligence. He has worked extensively on aspects of the social and economic impact of digital technologies, topics on which he is involved in management training and consulting. He regularly publishes in Italy for Econopoly (IlSole24Ore) and Agenda Digitale, and in Switzerland for Innovando News, and is a frequent keynote speaker or moderator at conferences and debates.

Claudio Sarra, Università degli Studi di Padova
Bio: Prof. Claudio Sarra is currently an Associate Professor, qualified as a Full Professor in the field of Legal Philosophy (GIUR 17/A), at the University of Padua. He holds a Law degree with honors, and a Ph.D. in Legal Philosophy. He regularly engages in international research and teaching activities. His professional work spans legal practice, particularly in civil and commercial law, and roles as a corporate legal consultant in technology law and data protection. Currently, he directs the second-level Master’s program in Bioethics and Biolaw at the University of Padua and serves as the Deputy Director of the Master’s program in Metaverse and Legal Informatics. He is a member of several scientific committees and international affiliations, including the Italian Society of Legal Philosophy and the Central and Eastern European Network of Jurisprudence (CEENJ), as well as the Ethics Committee of the Interdepartmental Center for Human Inspired Technology (HIT) and the Scientific Committee of the Center for Connectivity and Territory V-SIX. His current research areas include data protection, artificial intelligence and law, AI ethics, the philosophy of legal language, and legal epistemology. His recent publications include monographs and articles on topics such as datafication, the ethical and legal challenges of AI, and the rhetorical analysis of legal discourse.

Stefano Micocci, Strategy and Development, FBK,
Bio: Stefano Micocci achieved a master's degree in electronic engineering from the University of Bologna in 1994. He works at Fondazione Bruno Kessler where he is a member of the office of the Head of Strategy and Development. At FBK, he was CTO in the Digital Health and Wellbeing department where he was responsible for the development of the Electronic Health Record of the Autonomous Province of Trento. He was eHealth expert for the Minister for Technological Innovation and Digital Transition, with the role of CTO for the implementation of mission 6 (health) of the National Recovery and Resilience Plan. He also worked for an in-house company of the Emilia Romagna Region where he was responsible for the development of the Electronic Health Record of the Emilia-Romagna Region. He participated in ministerial technical groups and is a member of the board of the standardisation body HL7 Italia.

Dr. Gabriella Scipione, Head of HPC Data Management and Data Analytics, CINECA,
Bio: Ha conseguito il dottorato di ricerca in Fisica presso l'Università di Bologna e attualmente dirige la divisione “HPC Data Management and Data Analytics” del dipartimento di High Performance Computing del CINECA. Le sue attività si concentrano su High Performance Computing (HPC), sulla gestione dei big data in cloud computing, sull'analisi dei dati sfruttando le risorse HPC e sulla creazione di applicazioni personalizzate per la visualizzazione dei dati scientifici. È stata responsabile tecnico e coordinatore di diversi progetti scientifici europei HPC in settori quali il meteo, il clima, l'ambiente e le scienze della vita. È stata coordinatrice della proposta del supercomputer Leonardo EuroHPC, la candidatura dell'Italia a ospitare il sistema EuroHPC pre-exascale presso il Polo Tecnologico di Bologna. Ha un ruolo attivo nelle attività di Leonardo HPC del CINECA. È PI in diversi progetti e contratti relativi a Destination Earth e ad altri gemelli digitali come GLORI (Global to Regional ICON Digital Twin). Recentemente ha iniziato a coordinare il team del CINECA che fornisce supporto HPC alle comunità AI e LLM e coordina le attività dell'AI for Science. Dal 2017 al 2023 ha partecipato a PRACE (Partnership for Advanced Computing in Europe), ricoprendo il ruolo di Council Delegate per l'Italia. Dal 2024 è membro dell’Infrastructural Advisory Board di EuroHPC.

Data management and analysis in complex scenarios and infrastructures

Abstract

In 2023, the Intesa Sanpaolo Innovation Center (ISPIC) advanced its R&D labs to focus on Frontier Research, integrating expertise in AI, Neuroscience, Robotics, and Climate Change. This initiative supports applied research projects and innovation hubs across Italy, addressing a data-intensive and complex landscape.

The 2024 reorganization emphasized cross-disciplinary collaboration, incorporating sectors such as Social and Art, Culture and Heritage with a strong focus on sustainability and impact. The Innovation Hubs are the primary instruments to achieve these goals, fostering high-impact ESG initiatives through partnerships between the ISP Group and external entities. These hubs prioritize applied research, technology transfer, and innovative projects, engaging stakeholders in strategic decisions and funding. A key challenge is developing systems that effectively use state-of-the-art techniques to integrate and analyse heterogeneous data sources, particularly in the Federated Learning Health Hub, which focuses on Privacy-Preserving AI techniques in healthcare.

Short Bio

Laura Li Puma graduated from the University of Turin with a degree in Computer Science, specializing in Artificial Intelligence. She also obtained a master's degree in Telecommunications during her tenure at Telecom Italia Lab, where she worked in the Mobile Value-Added Service research field from 2000 to 2006.

With extensive experience in banking innovation, she has held roles in both IT and Organizational departments, ultimately gaining responsibility for Intesa Sanpaolo’s Big Data and IoT Office. In this role, she focused on enhancing expertise in monitoring business evolution, identifying emerging threats and opportunities from technological and regulatory changes, and leading Proof of Concept (PoC) developments.

Currently, she is part of the Intesa Sanpaolo Innovation Center in Turin, where she leads the Applied Research & Innovation Hubs—advanced R&D labs focused on Frontier Research that integrates expertise in AI, Neuroscience, Robotics, and Climate Change. In this role, she manages and coordinates relationships with scientific partners to tackle and develop business challenges both within and beyond the bank. By leveraging cutting-edge “applied data science” competencies, she collaborates with international researchers to address real-world use cases.

Crowdsourcing has become an important method for data collection, leveraging the power of distributed human intelligence to perform tasks that are often tedious for individuals or challenging for automated systems. Despite its advantages, the complexity of designing and managing crowdsourcing tasks can be a significant barrier, especially for those without extensive technical skills. This tutorial introduces the Crowd_Frame framework, a tool that simplifies the creation and deployment of complex crowdsourcing tasks, offering a user-friendly and flexible solution for researchers and practitioners.

Crowd_Frame is designed to address common challenges in crowdsourcing such as task design, worker management, and data quality. It provides a robust yet simple interface for setting up tasks, integrating components like task generators, user interaction skeletons, customized search engines, and data logging facilities. This integration allows users to configure and deploy a variety of task types easily and effectively, from simple surveys to complex data annotation tasks.

The Generator component of Crowd_Frame allows users to configure the structural elements of a task, tailoring everything from input forms to data validation rules. The Skeleton handles the user interaction, ensuring a seamless experience for the workers. Search Engine integration facilitates the inclusion of web-based research into tasks, providing workers with the tools to perform informed and accurate data entry. Finally, the Logger captures detailed metrics on worker behavior, providing data for analyzing task efficiency and worker engagement.

During the tutorial, participants will engage in hands-on exercises to experience the full capabilities of Crowd_Frame. They will learn how to design a sample task, to configure it, customize it, and how to deploy it.

Crowd_Frame builds on the foundational concepts of crowdsourcing discussed in works like “The Rise of Crowdsourcing” by Howe, which outlines the early developments and applications of crowdsourcing in multiple settings. Further, it addresses the technical complexities faced by researchers and practitioners when deploying crowdsourcing tasks on commercial micro task platforms.

Presenters

Michael Soprano is a Postdoctoral Research Fellow (PostDoc) at the University of Udine, Italy. His research interests center around Information Retrieval and Crowdsourcing. Particularly, he studies crowdsourcing-based approaches to address the ever-increasing problem of misinformation spread. He has published papers in top-ranked (CORE A* or CORE A) and selective conferences such as SIGIR, CIKM, and WSDM, as well as in top-tier journals (Q1 or Q2) such as IP&M, PAUC, TOIS, IS, and JDIQ. Additionally, he has visited and collaborated with the Human-Centered Data Analytics (HCDA) Group at the Centrum Wiskunde & Informatica in Amsterdam. Furthermore, he has collaborated with multiple universities across the globe, including RMIT University, The University of Queensland, University of Copenhagen, and Delft University of Technology. As a result of his work, he received two student grants, which provided him with free registration to participate in the 15th ACM International WSDM Conference and the 29th ACM International CIKM Conference.

Kevin Roitero is a Tenure-Track Assistant Professor (RTDb) at the University of Udine, Italy. His research interests include Artificial Intelligence, Natural Language Processing, Information Retrieval, and Crowdsourcing. He visited and collaborated with multiple top universities across the globe (The University of Sheffield, RMIT University, The University of Queensland, Pioneer Center for Artificial Intelligence & University of Copenhagen) as well as with leading industry partners (Spotify), publishing papers in top ranked (CORE A*) and selective conferences such as SIGIR, WSDM, WWW, CIKM, HCOMP, and in top-tier journals (highly ranked or Q1) such as TDKE, JDIQ, IRJ, and IP&M. As result of the his work, he received multiple grants and awards, including the participation in the 7th edition of the prestigious Heidelberg Laureate Forum (top 200 young researchers in mathematics and computer science), the “con.Scienze2020” prize for the best Ph.D. thesis in Computer Science discussed in 2020 in Italy, and multiple Best Paper Awards.

Stefano Mizzaro is Full Professor of Information processing systems at the University of Udine (Udine, Italy), where he teaches courses on Information Retrieval, Social Computing, and Artificial Intelligence. His main research interests are information retrieval evaluation, crowdsourcing, misinformation detection, mobile devices and systems, and scholarly publishing and peer review. According to Google Scholar1 he has more than 200 publications, almost 5000 citations, and an h-index of 33. He has published extensively in conferences such as SIGIR, CIKM, WWW, WSDM, ACL, ICTIR, HCOMP, and ECIR, as well as in journals like TOIS, TKDE, IP&M, IRJ, and JASIST. He has received several awards for best papers. His research has been funded by projects at regional, national, and European levels, as well as by private companies like Google and Facebook.

In the age of information, effective data management is crucial, particularly in academic institutions where diverse data integration can yield significant insights and advancements. This paper presents a pioneering graph database developed for the University of Naples Federico II, focusing on data aggregation and enhanced accessibility through an interactive dashboard. The tool helps university management monitor and analyze funded projects and grants from the European Union and Italian government, simplifying complex data for strategic decision-making and financial planning. It allows real-time tracking of projects, promoting interdisciplinary collaboration, aligning research impact with trends, and facilitating informed decisions.

Conference attendees can engage with the interactive dashboard, experiencing its robust data integration capabilities and visual representation of complex datasets.

Presenters

Gian Marco Orlando is currently a Graduate Research Fellow at the University of Naples Federico II. He earned his Master's Degree in Computer Engineering from the University of Naples Federico II, graduating with honors. His thesis highlights his expertise in the intersection of Artificial Intelligence and Social Network Analysis. He has also been an alumnus of the Apple Developer Academy, a collaboration between the University of Naples Federico II and Apple, where he received a diploma focusing on software development and app design. His research interests lie in Social Network Analysis, Agent-Based Modeling and Big Data Analytics.

Diego Russo is presently a Graduate Research Fellow at the University of Naples Federico II, where he also completed his Master's degree in Computer Engineering. His Master's thesis explores Recommender Systems and Big Data, demonstrating his advanced knowledge and research specialization in these innovative fields of Computer Engineering.

Valerio La Gatta is currently a Postdoctorate Scholar at Northwestern University, USA, in the Department of Computer Science. He earned his Ph.D. in Information Technology and Electrical Engineering (ITEE) at the University of Naples Federico II, focusing on the adoption of Artificial Intelligence techniques to address disinformation detection and intervention. His main research interests fall at the intersection of Social Network Analysis, Computational Social Science and Artificial Intelligence.

Antonio Galli is currently an Assistant Professor (non-tenure track RTD-A) at the University of Naples Federico II, in the Department of Electrical Engineering and Information Technology. He earned his Ph.D. in Technology, Innovation, and Management (TIM) at the University of Bergamo and the University of Naples Federico II, focusing on artificial intelligence applications in Industry 4.0. He graduated with honors in Computer Science. His main research interests are in the areas of deep learning and big data analytics.

Vincenzo Moscato is a Full Professor at the Electrical Engineering and Information Technology Department of University of Naples 'Federico II'. He received the Ph.D. degree in Computer Science from the same University by defending the thesis: 'Indexing Techniques for Image and Video Databases: an approach based on Animate Vision Paradigm'. He is one of the leaders of PICUS (Pattern and Intelligence Computation for mUltimedia Systems) departmental research groups and a member of the Big Data and Artificial Intelligence national laboratories within the Consorzio Interuniversitario Nazionale per l'Informatica (CINI). His research activities lay in the area of Multimedia, Big Data, Artificial Intelligence and Social Network Analysis. He was involved in many national and international research projects and coordinated as principal investigator some of them. He was in the program committees of numerous international conferences and in the editorial boards of several important journals. Finally, he was an author of about 200 publications on international journal, conference proceedings and book chapters.

Nowadays, sensors are deployed in public areas for instance with the scope of monitoring people behaviors for safety reasons. Such sensors (e.g., video cameras and air pollution sensors) gather data to be processed by services hosted on a distributed ICT infrastructure.

Safety-related services aim to address specific goal such as criminal behavior detection, critical massive people movement prediction, critical pollution concentration detection, and traffic anomalies prediction. They have to be deployed in the ICT infrastructure in order to satisfy specific QoS requirements.

Such infrastructure is therefore becoming a crucial asset to guarantee public safety, and its cybersecurity is of paramount importance. A simple denial of service at infrastructure level for instance, can have a tremendous impact on the ability to detect safety critical event.

Guarantee cybersecurity of critical public safety distributed ICT infrastructure poses a number of challenges including the difficulties of verifying weaknesses and vulnerabilities in capillary distributed and hybrid architecture. The trend of using 5G and Telco network to connect sensors and cover areas where wired connections are not available is even making the scenario more complex.

In this tutorial Moon Cloud, a cybersecurity monitoring framework is presented. It provides a framework for governing cybersecurity in a distributed architecture offering both an holistic view and details on how to address the detected cybersecurity issues. In addition the tutorial will also present MUSA a Framework for deploying services on distribute ICT infrastructure making them monitorable for security and privacy.

Presenters

Filippo Berto is a Research Assistant at the Department of Informatica “Giovanni degli Antoni”, University of Milan. He holds a doctoral degree in Computer Science (with honors) from the same university. His research focuses on Quality and Security Assurance for distributed systems, particularly Edge Cloud Continuum, cloud, IoT, and 5G networks. He develops methods and solutions to ensure advanced properties in these systems, such as reliability, privacy, and security. His work also involves the development of advanced network protocols for efficient content distribution, big data platforms, and data analysis pipelines. He is currently exploring systems based on Federated Learning for ensuring privacy and scalability on a large scale.

Marco Anisetti is an Full Professor at the Università degli Studi di Milano, Italy. His research interests are in the area of Computational Intelligence and its application to the design and evaluation of complex distributed systems and microservices. More in detail, he has been investigating innovative solutions for non-functional property assurance and certification in modern distributed systems. In this area, he defined a new scheme for continuous and incremental service security certification, based on distributed assurance evaluation architecture that percolated in his Moon Cloud spin-off. He is currently investigating the application of Big Data technologies for i) the computation of security assurance metrics including behavioral-based analytics, ii) the design of security and privacy-aware applications, iii) computing trustworthiness and assurance metrics in Edge Cloud Continuum, and for AI models. He participated in more than 10 EU projects including FP7 ASSET4SOA and FP7 CUMULUS, H2020 EVOTION, H2020 CONCORDIA, H2020 IMPETUS, H2020 CounteR to name but a few. The results of research activities have been published in more than 140 papers in international conference/workshop proceedings, journals, and chapters in books.

Claudio A. Ardagna is Full Professor at Dipartimento di Informatica, Università degli Studi di Milano, Italy. His research interests include information security and privacy in cloud and IoT, security assurance and compliance, and Big Data analytics. Within these areas, he has published 130+ contributions in international journals, conference/workshop proceedings, and chapters in international books. He is a co-author of the book “Open Source Systems Security Certification” published by Springer. He has been visiting researcher at Khalifa University and George Mason University.

The tutorial is organised in three presentations on the contributions of statistics and machine learning to the study of complex data. The contributions that will be presented at the tutorial are as follows:

pivmet: an R package proposing pivotal methods for consensus clustering and mixture modeling (Leonardo Egidi, Roberta Pappadà, Francesco Pauli, and Nicola Torelli)

We introduce the R package pivmet, a so6ware that performs different pivotal methods for identifying, extracting, and using the so-called pivotal units that are chosen from a partition of data points to represent the groups to which they belong. Such units turn out be very useful in both unsupervised and supervised learning frameworks such as clustering, classification and mixture modeling.

More specifically, applications of pivotal methods include, among the others: a Markov-Chain Monte Carlo (MCMC) relabelling procedure to deal with the well-known label-switching problem occurring during Bayesian estimation of mixture models; model-based clustering through sparse finite mixture models (SFMM)ti consensus clustering, which may allow to improve classical clustering techniques—e.g. the classical k-means—via a careful seeding; and Dirichlet process mixture models (DPMM) in Bayesian nonparametrics.

Treatment effects evaluation in observational studies: a propensity score method based on Bayesian Networks (Federica Cugnata, Paola Vicard, Paola M. V. Rancoita, Alberto Briganti, Fulvia Mecatti, Clelia Di Serio and Pier Luigi Conti)

In medical research, the evaluation of treatment effects on outcomes using observational study data has become increasingly important. One of the main challenges in such studies is the presence of relevant differences between treatment groups, o6en due to biases in treatment decisions. These differences can lead to biased estimates of treatment effects. Methods based on propensity scores are widely adopted to account for this source of bias. However, its use requires a careful modelization of the dependence relationships of the treatment on the covariates. In this work, we propose to estimate the propensity score by using Bayesian Networks and, based on this, we develop an inferential methodology to evaluate the treatments effect.

A new proposed data-driven index to monitor conflicts (Luca Macis, Marco Tagliapietra, Elena Siletti, Paola Pisano)

In this work, we propose a Violence Index (VI), as a comprehensive indicator of violence related to wars, conflicts, and disorders across different countries worldwide. This index is defined by mashing up different data sources: big data represented by the temporal progression of ACLED (Armed Conflict Location and Event Data) variables, and an ad hoc dataset defined by our team documenting wars, armed conflicts, civil wars and violent demonstrations since 2010. The purpose is to encapsulate the intensity and impact of such unrest events in a single measure, the VI, to give a simpler, up-to-date, and manageable tool to practitioners and policymakers, for both prevention and strategic planning to let them behave better in future tragic scenarios.

Presenters

Leonardo Egidi is assistant professor of Statistics at Department of Economics, Business, Mathematics, and Statistics of the University of Trieste. His main research areas cover theoretical and applied Bayesian statistics, with a particular focus on robust prior elicitation and hierarchical models, sports modelling and clustering methods. He is a proficient R package developer -pivmet, footBayes, clc- and a Stan user. He has open collaborations with Italian and foreign institutions, such as Athens University of Economics and Buisiness (AUEB), Columbia University, University of Pavia, CaQolica University of Milan. He teaches master-level and phd-level courses, and he is a member of the ADSAI PhD college (University of Trieste).

Federica Cugnata is Assistant Professor in Statistics at the Faculty of Psychology (Vita-Salute San Raffaele University), and involved in the research activities at CUSSB (University Centre for Statistics for the Biomedical Sciences). She participated in numerous research projects across different fields and her main research interests focus on graphical models, statistical models for ordinal data, linear and nonlinear mixed-effect models, and multivariate statistics.

Elena Siletti is an RTD-B researcher in Statistics at the Department of Economics and Statistics 'Cognetti de Martiis0', University of Turin. Member of the management committee of the HighESt laboratory at the University of Turin for theoretical and applied research in data analysis and artificial intelligence. Elena has participated in numerous research projects in different areas of high scientific impact, including multidisciplinary, using innovative applications and methodologies.