Social Networks and Big Data Methods (Summer 2026): Difference between revisions

Hqiu (talk | contribs)
Hqiu (talk | contribs)
 
(6 intermediate revisions by the same user not shown)
Line 16: Line 16:


Evaluation will be based on attendance (10%), class participation (10%), presentation (30%), and final project (50%).
Evaluation will be based on attendance (10%), class participation (10%), presentation (30%), and final project (50%).
Paper presentation requirements:
1. At least 12 mins;
2. Clear statement of the research questions in the papers;
3. Focus on methods and results.
Final project requirements: replicate a paper related to social network analysis:
1. Single column;
2. 10pt font, 1.5x line spacing;
3. You can have as many images and tables as you want;
4. 10 pages;
5. Focus on the methods and results;
6. For all networks, present their basic descriptions (size, density, degree distributions, number of largest components, size of largest components, etc)


Slides: https://github.com/sophieball/social_network
Slides: https://github.com/sophieball/social_network
Line 24: Line 46:


Software: Python Networkx, Jupyter notebook, pandas
Software: Python Networkx, Jupyter notebook, pandas
Example: https://github.com/sophieball/social_network/tree/main/code


Tutorial: http://networkx.org/documentation/stable/tutorial.html
Tutorial: http://networkx.org/documentation/stable/tutorial.html
==Ideas for Final Project==
1. Do Women Suffer from  Network Closure? The Moderating Effect of Social Capital on Gender Inequality  in a Project-Based Labor Market, 1929 to 2010
Data:https://www.kaggle.com/datasets/rishabjadhav/imdb-actors-and-movies
2. Online extremism and the communities that sustain it: Detecting the ISIS supporting community on Twitter
3. Identifying Social Roles in reddit Using Network Structure


==Schedule (Tentative)==
==Schedule (Tentative)==
Line 42: Line 76:
| align="right" | 22.04.2026 (14:00-16:00)
| align="right" | 22.04.2026 (14:00-16:00)
| Random networks; Edges vs social ties; sign up for presentation
| Random networks; Edges vs social ties; sign up for presentation
|
|[https://github.com/sophieball/social_network/blob/main/w2_1_random_networks.pdf slide3]
|  
|  
|-
|-
| align="right" | 29.04.2026 (14:00-16:00)
| align="right" | 29.04.2026 (14:00-16:00)
| Triads and structural balance; From social processes to graphs
| Triads and structural balance; From social processes to graphs
|
|[https://github.com/sophieball/social_network/blob/main/w2_2_edge_vs_social_ties.pdf slide4] [https://github.com/sophieball/social_network/blob/main/w3_1_triads_and_structural_balance.pdf slide5]
|  
|  
|-
|-
| align="right" | 06.05.2026 (14:00-16:00)
| align="right" | 06.05.2026 (14:00-16:00)
| Homophily and degree correlation (parts 1 & 2)
| Cancelled - Sophie is sick.
|
|
| LLMs generate structurally realistic social networks but overestimate political homophily
|  
|-
|-
| align="right" | 13.05.2026 (14:00-16:00)
| align="right" | 13.05.2026 (14:00-16:00)
| Power and centrality in social networks; social exchange
| Power and centrality in social networks; social exchange
|
|[https://github.com/sophieball/social_network/blob/main/w5_1_power_and_centrality_in_social_networks.pdf slide6] [https://github.com/sophieball/social_network/blob/main/w5_2_power_and_centrality_in_exchange.pdf slide7]
| Investigating Centrality Measures in Social Networks with Community Structure
| Investigating Centrality Measures in Social Networks with Community Structure
|-
|-
Line 66: Line 100:
|-
|-
| align="right" | 27.05.2026 (14:00-16:00)
| align="right" | 27.05.2026 (14:00-16:00)
| Affiliations and overlapping subgroups
| Affiliations and overlapping subgroups, SUBMIT FINAL REPORT IDEA
|
|
| Overlapping community and entropy of neighborhood information for identifying influential nodes in complex networks
| Overlapping community and entropy of neighborhood information for identifying influential nodes in complex networks
|-
|-
| align="right" | 03.06.2026 (14:00-16:00)
| align="right" | 03.06.2026 (14:00-16:00)
| TBD
| Homophily and degree correlation (parts 1 & 2)
|
|
|  
| LLMs generate structurally realistic social networks but overestimate political homophily
|-
|-
| align="right" | 10.06.2026 (14:00-16:00)
| align="right" | 10.06.2026 (14:00-16:00)