Abstract: In these lectures I will introduce some key topics related to the analysis and prediction of human behaviour and social dynamics for urban data science. In particular, I will present two different types of approaches: first of all, I will discuss how it is possible to use the digital traces (mobile data, social media, etc.) that we leave in our online and offline daily lives in order to model human behaviour at a scale and granularity that were not possible just a few decades ago. In the second part of the lecture, I will give an overview of the emerging computational paradigms based on Machine Learning (in particular Reinforcement Learning) for modelling individual responses and understanding key emerging aspects of social behaviour such as cooperation and coordination. I will discuss some case studies with a focus on urban data science. I will give an overview of the open challenges in this fascinating area.

Speaker: Mirco Musolesi is a Reader in Data Science at University College London and a Turing Fellow at the Alan Turing Institute, the UK national institute for Data Science and Artificial Intelligence. At UCL he leads the Intelligent Social Systems Lab. He held research and teaching positions at Dartmouth, Cambridge, St Andrews and Birmingham. He is a computer scientist with a strong interest in sensing, modelling, understanding and predicting human behaviour and social dynamics in space and time, at different scales, using the "digital traces" we generate daily in our online and offline lives. He is interested in developing mathematical and computational models as well as implementing real-world systems based on them. This work has applications in a variety of domains, such as intelligent systems design, ubiquitous computing, digital health, security&privacy, and data science for social good. More details about his research profile can be found at: http://www.mircomusolesi.net

Abstract: Measuring dynamics and phenomena occurring in urban space is a fundamental step in the process of understanding cities. The recent availability of large amounts of urban data (also known as ‘big data’), often containing human mobility information at a higher spatiotemporal resolution, represents an unprecedented opportunity for urban researchers of exploring and unveiling new characteristics of urban space, quantitatively validating established theories, and ultimately strengthening the overall understanding of spatial dynamics happening in cities. Numerous studies have explored and employed these data, applying multiple analysis techniques adopted from quantitative methods and data science to different urban contexts and obtaining a variety of outcomes. This talk presents an overview of current research related to urban analysis and urban computing: it introduces and discusses the definition of big data, presents the most common types of data employed in the literature, and illustrates the applications of these data sets and techniques to a number of different case studies across cities. It also discusses the limitations present in current research, particularly in terms of data representativeness and determinism, and the advantages and opportunities that exploiting these data represent for enhancing the understanding of cities and improving future urban developments.

Speaker: Patrizia Sulis is a spatial data scientist and urban analyst. She holds a PhD in Spatial Data Analysis from University College London and an MSc in Urbanism from the Delft University of Technology. She worked as Teaching Assistant in the courses of Smart Cities and Urban Data Analysis at University College London, and as GIS Specialist and Research Assistant for the Centre for Advanced Spatial Analysis (UCL), the Bartlett School of Planning (UCL) and the Delft University of Technology. Her doctoral research, supervised by Dr Ed Manley and Prof. Mike Batty, focused on the application of data science and quantitative methods to urbanism and urban analysis. She developed a computational approach investigating human mobility patterns and the spatiotemporal variation of urban vitality in cities, using a variety of ML techniques and extensive urban data sets sourced from public transport, social media and open sources.

Abstract: Smart cities rely on many different data sources. Several of these data can however be considered as personal data, because they are related to an identified or identifiable natural person. It is therefore fundamental, in the development of any smart cities service or device, to apply the principles of EU Regulation 2016/679 (General Data Protection Regulation - GDPR) from the very start. Data protection issues cannot be an afterthought, but they have to be integrated by design, as art. 25 of the GDPR clearly states. The talk will therefore summarize the key points of the EU General Data Protection Regulation, with regard to smart cities.

Speaker: Giovanni Battista Gallus, LL.M., Ph.D., parter of Array Law, Data Protection Officer for several public and private entities.

Abstract: Urban computing systems must acquire, integrate, analyze, and present big and heterogeneous data generated by a diversity of sources in urban spaces. In this talk, we will describe the design of such a system using open standards. Examples of practical realizations in the context of the TDM project will be discussed.

Speaker: Enrico Gobbetti is the director of Visual Computing at the Center for Advanced Studies, Research, and Development in Sardinia (CRS4), Italy. He holds an Engineering degree (1989) and a Ph.D. degree (1993) in Computer Science from the Swiss Federal Institute of Technology in Lausanne (EPFL). His main research interests span many areas of visual computing, with emphasis on scalable technology for acquisition, storage, processing, distribution, and interactive exploration of complex objects. Systems based on these technologies have been used in as diverse real-world applications as internet geoviewing, scientific data analysis, surgical training, and cultural heritage study and dissemination. Enrico has (co-)authored over 200 papers in visualization and computer graphics, six of which received best paper awards. He regularly serves the scientific community through participation in editorial boards, conference committes, and working groups, as well as through the organization and chairing of conferences. He is a Fellow of Eurographics.

Speaker: Gianluigi Zanetti is the director of Data-intensive Computing at the Center for Advanced Studies, Research, and Development in Sardinia (CRS4), Italy. He holds a degree in Physics from the University of Bologna (1984) and a Ph.D. in Physics from The University of Chicago (1988). Before joining CRS4, he has conducted research and teaching activities at Princeton University, Los Alamos National Lab, École Normale Supérieure (Paris) and other institutions. His current research interests focus on the development of scalable mechanisms to support data-intensive research with applications in the biomedical sciences and urban computing.