Foto 7

Dr. Stefano Cresci, IIT CNR, Pisa, "Credibility assessment in social media with a focus on social bot detection", 25,26,27,28 January 2022

Hours:
12 hours (3 credits)

Room:

Aula Riunioni del Dipartimento di Ingegneria dell’Informazione, Via G. Caruso 16, Pisa – Ground Floor

To register to the course, click here

Short Abstract:

We live in a time of automated information warfare, where hordes of bots (automated software programs) rampage in our online social ecosystems. These content-polluting bots are employed by both malicious actors aiming to spread misinformation, as well as by traditional news outlets fighting for readers’ attention.
In this course, we introduce and we experiment with the fundamentals of social media crawling and analysis, we discuss the issues related to information credibility in social media, and we investigate the role of social bots in the spread of high- and low-quality (e.g., fake news) information. We also introduce the task and the challenges of social bot detection. Then, we report on social media platforms capabilities of detecting bots, and on human performance in discriminating between legitimate and bot accounts. Finally, we thoroughly discuss different machine learning and AI approaches to the automatic detection of social bots (e.g., network-, content-, and behavior-based), highlighting those that currently represent the most promising ones. The course also includes hands-on sessions in Python, where participants will write scripts for collecting live data from Twitter and will learn to perform bot detection. The course also covers points such as available datasets, software, and APIs for supporting the study of information credibility and the detection of social bots.

Course Contents in brief:

  1. Social media and information credibility (e.g., fake news, coordinated inauthentic behavior);
  2. Social bots and their role in the spread of low-quality information;
  3. Early (i.e., supervised) and recent (i.e., unsupervised, group-based) machine learning/AI approaches to the detection of social bots;
  4. A look into the future of information credibility and social bot detection;
  5. Data collection from Twitter and bot detection experiments in Python.

Schedule:

  1. Day1 – 25 January 2022, 9:00 - 12:00
  2. Day2 – 26 January 2022, 9:00 - 12:00
  3. Day3 – 27 January 2022, 9:00 - 12:00
  4. Day4 – 28 January 2022, 9:00 - 12:00