Course Overview

During the past years, data science played a pivotal role in many social issues such as the spreading of (mis)information online, echo chambers, bot detection, mobility patterns, and resilience.

The course will introduce advanced topics of networks science and diffusion models and, thanks to the complex systems approach, we will extract knowledge from data by exploiting a complexity-driven (e.g., dynamic processes on complex networks) approach.

In the first part we will explore the foundational aspects and advanced topics of complex systems (multilayer networks, percolation, time-varying graphs), while in the second part, we will apply those concepts to state-of-the-art cases on massive datasets, ranging from the effect of feed algorithms on social dynamics up to patterns of human mobility, passing through information operations, and bot detection.

List of Topics

1. Introduction

• Introduction to the course

• A complete example of data-driven modeling of complex systems: the case of misinformation diffusion

2. Advanced Concepts in Complex Systems

• Recall of Network Science concepts (stochastic network models and metrics)

• Implementation of the small world effect

• Preferential Attachment and other generative models

• Multilayer networks

3. Networks from Data and Processes

• Dive into R-igraph

• Accessing massive social data online (Facebook, Twitter, Youtube, Reddit)

• Spreading processes on different type of networks

• Voter Model on different type of networks and Bounded Confidence Model on different types of networks

4. A case study from massive datasets to insights: Misinformation

• The spreading of misinformation online (https://www.pnas.org/doi/10.1073/pnas.1517441113)

• Cyber threats: Information and psychological operations (https://www.sciencedirect.com/science/article/pii/S0167923622000902)

• Information operations and the detection of Social Bots (https://dl.acm.org/doi/pdf/10.1145/3292522.3326015)

5. Dynamic Networks and Percolation Theory in Human Mobility

• Human mobility during the pandemic (using Facebook Data) (https://www.pnas.org/doi/abs/10.1073/pnas.2007658117)

• Self-healing networks (https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0087986)

• Modeling resilience of complex systems (https://www.nature.com/articles/s41598-021-92399-2)

6. Social Media Algorithms and Social Dynamics

• Capturing the polarizing effect of feed algorithms (https://www.pnas.org/doi/abs/10.1073/pnas.2023301118)

• Memes as a language: Unraveling the evolution of memes complexity (https://www.nature.com/articles/s41598-021-99468-6)

• The evolution of toxicity over platforms (https://www.sciencedirect.com/science/article/pii/S2352250X22001282)

Timetable

Tuesday 11.00 - 13.00 Room 2 Viale Ariosto 25, Rome
Friday 11.00 - 14.00 Room 2 Viale Ariosto 25, Rome

Materials:

All the materials can be reached by registering to the course's page on classroom (code: xfyrmjq).

Contacts

walter.quattrociocchi@uniroma1.it