Friendly Borders: Graph algorithms reveal Eurovision voting patterns
summary
Claims neighborly favoritism is real and quantifies it using 2008–2024 Eurovision voting data, ingesting a community-maintained JSON dataset, building graphs in NetworkX, running Louvain modularity repeatedly, creating a co‑association consensus via NetworkX + Monte Carlo consensus, and visualizing with Plotly. Results: stable regional clusters (Yugoslavia 2.0, Mediterranean, Iberia), televote shows stronger neighbor bias, juries favor Sweden, betweenness centrality flags Ukraine and Israel as public vote hubs, Italy is the big‑five outlier that consistently scores well.
AI-written summary · about
chapters (18)
- 0:00Introduction and speaker intro
- 1:09What is a graph in math
- 3:38Graph theory problems overview
- 6:09Business cases for graph algorithms
- 8:57Eurovision Song Contest basics
- 11:51Thesis and research questions
- 12:33Community detection algorithms
- 14:19Solution and data preparation
- 17:26Data preprocessing and tools
- 19:15Community detection results
- 21:01Analyzing the big five countries
- 22:47Jury vs public voting bias
- 25:15Notes and final thoughts
- 27:22Conclusion and Q&A intro
- 28:25Q&A on social network scraping
- 31:11Q&A on future data extrapolation
- 32:16Q&A on combining political data
- 33:44Q&A on sharing results
