Data-driven Network Science

About

Systems composed of elements in interaction can naturally be represented as networks. There is a long tradition in social sciences concerning social network analysis, but over the last 20 years alternative network analysis methods have been developed under a complex systems approach.

The course will present in detail a range of methods to extract information from real-world systems. Statistical Analysis of Networks will introduce the students to network summaries and network models, aiming at understanding how the structure of a network impacts its function, often associated to dynamical processes running on its nodes. As a motivating example, consider information diffusion in social networks.

Important questions would include:

• Do social networks exhibit significant structural patterns?
• Which mechanisms are likely to explain the formation of these patterns?
• How do these patterns impact the diffusion of information between agents?

Key Themes

• Network statistics
• Models of networks
• Dynamics on networks

Learning Objectives

At the end of this course students will understand state-of-the-art algorithms for network mining know how to formulate scientific questions related to network data know how to perform numerical experiments and statistical tests be able to report on the analysis of network data in a critical way.