Corso di laurea magistrale in Informatica

The purpose of the course is threefold. First, students will learn the fundamentals of network science (NS), including techniques for analyzing complex networks, and basic models for studying network dynamics (e.g., social contagion, viral phenomena, the spread of epidemics, etc.). Second, they will learn how to analyze datasets that can be represented in the form of networks, through the computational study of graph structure and the dynamic processes that occur on it.
Finally, they will be taught the basic principles of information visualization and in particular how to handle the visual analysis of complex information representable in the form of a graph and beyond.

This teaching contributes to the educational objectives in the area of computer networks and systems, as well as information processing systems in the Computer Science Master's degree program.

KNOWLEDGE AND CAPACITY TO UNDERSTAND: Students will deepen their knowledge of basic algorithms and data structures for the analysis of complex networks and for the visual representation of scientific and complex information.

ABILITY TO APPLY KNOWLEDGE AND UNDERSTANDING: Students will be able to analyze data, including visual processes, presented in the form of a network, using tools from public domains.

JUDGMENTAL AUTONOMY: Students will be able to independently make an assessment of the most appropriate techniques for solving a generic network science problem and complex information visualization.

COMMUNICATION SKILLS: Students will learn to communicate and justify the use of the most appropriate techniques for solving a given network science and information visualization problem.

LEARNING SKILLS: Students will be able to learn independently how to refine the basic techniques they will learn in the course and how to best use the available tools.

Complex Networks and "Network Analysis"

• Introduction to complex networks
• Recap of graph theory
• Strong and weak ties
• Homophily
• Complex Network Analysis (CNA) with Python;
• Centralities, Robustness, Structural Holes, Graph Partitioning
• Network Models, Random Networks and Small World Networks
• Preferential Models
• Communities
• Signed Networks
• Directed and Weighted Networks
• The Structure of the Web
• Link Analysis and Web Search
• Information Cascades
• Cascading Behavior in Networks
• Network Effects
• Power Laws and Rich-Get-Richer Phenomena
• Small World and Search
• Epidemics; Epidemic spreading in heterogeneous networks
• Introduction to temporal graphs and hypergraphs
• Graph Representation Learning (Node embeddings, Graph Neural Networks, Generative Graph Models)

Information Visualization

• Basic Charts and Plots, Multivariate Data Visualization
• Principles of Perception, Color, Design, and Evaluation
• Text Data Visualization
• Interactivity and Animation
• Temporal Data Visualization
• Geospatial Data Visualization
• Hierarchical Data Visualization
• Network Data Visualization

Lectures that introduce theoretical concepts and laboratory exercises that apply them. In the lessons, theoretical topics are addressed through slide presentations, with examples and some questions to verify students' learning. In laboratory exercises, students are guided in the implementation of simple projects aimed at putting into practice the theoretical knowledge acquired.

The exam consists of three parts:

• Project I (30%): data visualization project from a dataset and according to the instructions given on Moodle
• Project II (20%): network analysis project on a dataset represented as a graph and according to the instructions given on Moodle
• Oral exam (50%): questions on the theory part and on the discussion of Project I (network analysis part) and of Project II (data visualization part)

A grade in the range [0,30] will be given for each of the above tests. The individual test will be considered passed if a score greater than or equal to 18/30 is obtained. The final grade will be calculated from the scores obtained in the individual tests by a weighted average according to the percentages shown above.

During teaching, students will interact with the lecturer in solving exercises and lab assignments, including through the Moodle platform.