Friendly Borders: Graph algorithms reveal Eurovision voting patterns

on YouTube ↗40m

" summary

Graph-based analysis uses Louvain with 1,000 Monte Carlo runs on Eurovision data (2008–present) built with NetworkX + pandas + Plotly to produce a consensus co-association graph; communities at resolution=1.1 reveal stable regional blocks (Balkans with Turkey, Iberia, Nordic/central Europe) confirming neighborly favoritism in public votes, while jury votes lack that neighbor bias and show Sweden as a high betweenness outlier; an auxiliary finding is Italy never in bottom five among the Big Five, so the methods produce reproducible, interpretable clusters.

AI-written summary · about

" chapters (36)

1. 0:00Introduction
2. 2:03What is a graph
3. 3:08Seven bridges and Eulerian paths
4. 4:34Shortest path problems
5. 5:40Traveling salesman basics
6. 6:44Centrality and communities
7. 7:46Graph algorithms use cases
8. 10:35Detecting astroturfing with graphs
9. 11:39Eurovision overview
10. 12:21Analyzing Eurovision with graphs
11. 13:26Eurovision voting system
12. 14:28Neighborly favoritism thesis
13. 15:56Louvain and Newman methods
14. 16:58Louvain non-overlapping method
15. 18:04Louvain algorithm randomness
16. 18:46NetworkX library
17. 19:28Eurovision dataset 2008-2024
18. 20:55Data preprocessing
19. 22:01Solution components
20. 22:44Building Eurovision graph
21. 23:27Monte Carlo simulation
22. 24:30Plotting communities
23. 24:52Analyzing communities
24. 25:36Resolution parameter impact
25. 26:18Final granular graph
26. 26:59Evaluating big five
27. 28:04Big five performance
28. 29:07Jury vs public bias
29. 29:52Cherry votes analysis
30. 30:54Betweenness centrality
31. 31:55Sweden’s jury outlier
32. 32:15Public votes geopolitics
33. 32:37Final thoughts
34. 34:23Graph databases
35. 36:08Main conclusions
36. 37:13Question and answer session

" related talks

• 28mNetwork analysis inside the Italian parliamentMario Migliaccio · PyCon Italia 2023
• 33mBuilding Data Visualisations in Python in MinutesKris Jenkins · GOTO Copenhagen 2025
• 42mVisualising ConnectionsNadieh Bremer · beyond tellerrand Berlin 2022
• 24mFootbal meets graphsBea Hernández · satRdays Paris 2019
• 46mGraph Algorithms: Predict Real-World BehaviorJennifer Reif · Jfokus 2019
• 30mAnalyzing COVID-19 Protest Movements: A Multidimensional Approach Using Geo-Social Media DataNefta Kanilmaz · PyCon Italia 2024
• 24mHands-On Network ScienceFarrelly, Gaba, Kalala Mutombo · PyData Global 2023
• 01h 03mCharting Collections of Social Media Connections with NodeXLMarc Smith · PASS Business Analytics Conference 2013