Recent years have witnessed an unprecedented availability of information on social, economic, and technological phenomena. Researchers, practitioners, and policymakers have nowadays access to huge datasets (the so-called “Big Data”) on people, companies and institutions, web and mobile devices, satellites, etc., at increasing speed and detail. Relational (big) data are also in a surge, thus documenting an increasing need to shed light on relationships among research and innovation actors. Network Science (NS), Bayesian Networks (BN), Machine Learning (ML) and Spatial Model (SM) are relatively new techniques able to enlarge our understanding of complex socio-technological systems, either by digging deeply into the data informative power (ML), or by increasing the understanding of the system relational dimension (NS, BN and SM). The training will provide participants with the essential tools for a correct application of some popular NS, BN, ML and SM methods in various STI contexts. In particular:
- ML techniques proves useful for factor importance detection, as well as for classification purposes in a model-free stance;
- NS and SM techniques are useful to identify and study structure and dynamics of large and complex STI communities.
The course foresees four modules (one on NS, one on BN, one on ML, and one on SM) with the aim of balancing theory and applications. Participants will run some exercises assigned by the instructor under his supervision.
This course directly contributes to the objectives of the FOSSR (Fostering Open Science in Social Science Research) project by fostering the development of new competences and knowledge that are essential for working with the Open Science Cloud being developed under FOSSR. By training participants in the use of NS, BN, ML, and SM techniques, the course supports the integration and analysis of diverse data sources within the Open Science Cloud, enabling researchers to generate more nuanced insights and promote data-driven decision-making in social science research.
The course offers an in-depth exploration of advanced statistical techniques and methodologies, focusing on Network Models, Bayesian Modeling, Machine Learning, and Spatial Models, specifically applied within the context of Research Infrastructures (RIs). Participants will learn to effectively engage with RIs and leverage their resources for research purposes, enhancing their understanding of data management and analysis in Science, Technology, and Innovation (STI) systems.
Innovative Aspects: This course integrates cutting-edge advancements in Network Science, Bayesian Networks, Machine Learning, and Spatial Models, tailored specifically for analyzing complex STI systems. The program emphasizes practical application using real-world datasets, making it a valuable resource for those working with research data.
Duration: The course spans four days, from February 24 to February 27, 2025, with daily sessions from 11:00 AM to 4:30 PM.
Target Audience: The course is designed for early-career researchers, data managers and technicians involved in the management, analysis and interpretation of research data, particularly within the social sciences.
Methodological Approach: The course is structured around interactive lectures, hands-on sessions using R, and dedicated Q&A segments, allowing participants to engage directly with experts and apply learned concepts in real-time.
Virtual Meeting and Working Space Platform: The course will be conducted online via a dedicated virtual learning platform, ensuring interactive participation and access to all necessary resources.
The course is conducted in English. All the materials will be published in English. It is designed for individuals who are or wish to be involved in creating, capturing, analysing, or generally managing research data within the social science disciplines.
Target audience: includes, but is not limited to, early-career researchers, researchers aspiring to advance their careers, technicians, data stewards, and data managers.
Deadline for application: 30th January 2025. Notification of acceptance by Feb. 7th, 2025.
Participation in the Winter School will be limited to a maximum of 40 attendees. If the number of registrations exceeds 40, a selection will be made based on the relevance of the educational or professional background, in relation to the course topics, as outlined in the submitted CV. In particular, basic requisites for admission will be: knowledge on basic principles of statistics; interest in STI studies.
Antonio Zinilli, researcher at CNR-IRCrES, he focuses on Network Science, dynamic processes on knowledge and innovation systems, and Text Mining. He is the coordinator of the IRCRES School in “Data Science: tools and methods for analyzing complex Science, Technology and Innovation (STI) systems”. He is a WP Leader in the PNRR project – FOSSR and is among the managers of the EFIL dataset, part of the European infrastructure RISIS for research and innovation studies.
Barbara Guardabascio is an Associate Professor in Economic Statistics and a member of the Business and Collective Intelligence Lab at the University of Perugia. Her research interests encompass economic forecasting, business cycle analysis, and big data, areas in which she has authored several articles in international journals.
Lorenzo Giammei, researcher at CNR-IRCrES. His studies focus on causal inference implementing approaches from both potential outcomes and causale Bayesian networks. He applied the mentioned methods on microeconomic research questions related to firms, gender gap and research productivity.
Giovanni Cerulli is senior researcher at CNR-IRCrES. His main areas of research are causal inference and machine learning. His major field of application is the economics of innovation although he has contributions also in other socio-economic areas. He is the coordinator of the PNRR project – FOSSR.
