Stefano Guarino

PhD Mathematics
Researcher at Institute for Applied Computing (IAC)
National Research Council of Italy

Areas of Expertise:
Data analysis and Security
Graph Algorithms

Short Bio:
Stefano Guarino earned his MSc and PhD in Mathematics at Roma Tre University. He is currently with the Istituto per le Applicazioni del Calcolo “Mauro Picone” of the Consiglio Nazionale delle Ricerche, where he he works on data analysis and security, with a recent focus on graphs and complex systems, addressing both methodological/algorithmic and implementation/technological aspects. He participated to several national and international research projects, including the H2020 Project “SOMA”, where he contributed to the definition of models and algorithms for the analysis of and the fight against disinformation on social media, and the EU ISEC Project “IANCIS”, where he worked on the extraction and correlation of semantics and topological properties of the Tor Dark Web. He is the PI of the MUR funded FISR Project “CARES”, which aims at defining a multi-layer graph-based model to simulate real-world person-to-person interactions and guide epidemic containment measures.

Sylllabus

Graph Algorithms
This course aims at providing students with an understanding of graph models and algorithms, taking a practical perspective focused on the analysis of social media data. The students will be given access to a dataset of Twitter posts (“tweets”) and, through examples and exercises, they will learn how to extract a network representation for information contained in the data and how to analyze the obtained graph to gain valuable knowledge about the social debate occurring on Twitter on the selected topic. The considered use-case will be instrumental to introduce graph theory and the empirical analysis of real-world networks – including, but not limited to, online social networks. The course will cover all the main topics of network theory, such as:

  • Network representation (adjacency matrix, adjacency lists, efficient representations for sparse graphs, …)
  • Measures and metrics (centrality, transitivity, assortativity, …)
  • Algorithms (shortest paths, DFS, BFS, Dijkstra, …)
  • Properties of real-world networks (small-world effect, scale-free networks, assortative mixing, …)
  • Graph models (random graphs, configuration model, preferential attachment, …)
  • Community structure (spectral clustering, modularity maximization, …)

The course will make use of Python and, in particular, of the Python interface to the igraph library.