DSOC 25A21 - SNA2: Social Network Analysis in the digital age

This course aims to introduce students to the field of Social Network Analysis (SNA). Social networks are ubiquitous nowadays, SNA emerged in the 1960s as a vibrant social science specialty trying to give substance to individuals, not through their inner psychological and demographic, or professional characteristics but through the relationships they entertain with their social environment. The first objective of this class will be to introduce the concepts and metrics designed and theorized by this specific stream of sociology and test how operative they still are in our connected environment. How useful are centrality or cohesion measures today? What can we learn about our current online world with concepts forged in the '60s and the '70s like homophily, transitivity, cohesion, diffusion processes? To do so, this course will examine the seminal papers in SNA. However, this intellectual journey will be complemented by a more hands-on approach, as half of the course will be devoted to teaching the students basic operations in Python such that they can collect data from digital social media platforms before modeling, measuring, and visualizing this data using recent network analysis libraries. We will put the ancient concepts of SNA to the test and assess how fruitful they are in understanding online interaction data. No prior coding experience is required as we will extensively use AI capacities (such as Gemini, directly available in Google Colab notebooks) to assist with coding. The class will alternate readings of historical sociology papers and more contemporaneous articles typical of the digital age mixing concepts from SNA in the larger realm of computational social sciences. Most classes will be split into three parts: the discussion around a scientific paper, a lecture about a new SNA-related concept, and a third part where students will be invited to experiment on their own laptops with the newly introduced concepts, metrics, or algorithms with empirical data.

Jean-Philippe COINTET

Séminaire

English

No previous knowledge in either graph theory or programming is required. The class will still require to read papers adopting formal approaches. So students should be ready to engage with such content.

Spring 2024-2025

You will be asked to read and present a scientific paper in line with the topic of the session. One paper per week will be presented by two or three students. Presentations will be evaluated (50% of the final grade) Additionally, a collective project will be conducted by students who are expected to build a network from social media platforms and analyze its structure (50% of the final grade) Student engaged during the semester can be rewarded by an extra point.