Hierarchical Clustering: Finding the awkward reunions

on YouTube ↗21m

" summary

Claims a hierarchical random graphs generative model infers multilevel community structure and reproduces right-skewed degree distributions, high clustering and short paths, then uses Metropolis Hastings MCMC to sample dendrograms and average posterior link probabilities. Implementation is partial in Python as pyrg pi hrg, applied to protein–protein interaction networks, yields higher AU statistics and R² versus similarity baselines and ranks a specific missing edge (A–D) as top candidate with ~0.385 average probability.

AI-written summary · about

" chapters (14)

1. 0:00Introduction
2. 1:44Speaker background
3. 3:10Predict missing links
4. 6:24Limitations of methods
5. 7:47Complex systems
6. 9:53Hierarchy emergence
7. 10:58Extracting hierarchies
8. 12:44Internal node probabilities
9. 14:31Sampling challenges
10. 15:55MCMC sampling
11. 16:57Link prediction ranking
12. 18:02Method advantages
13. 19:07Protein interaction app
14. 20:31Python package

" related talks

• 43mUsing network community clustering algorithms [...]Claudia Millán · PyData Barcelona 2017
• 35mClustering: A Guide for the PerplexedLeland McInnes, John Healy · PyData DC 2016
• 41mGeneral Network Models with Transitive StructuresSharmodeep Bhattacharyya · Statistics in the Big Data Era 2022
• 58mIntroduction to Biological Network Analysis I: Network Basics and PropertiesDonna Slonim · Algorithmic Challenges in Genomics Boot Camp
• 25mConstrained Rank Aggregation and Correlation ClusteringOlgica Milenkovic · Information Theory, Learning and Big Data (Simons Institute, 2015)
• 47mLearning Probabilistic Models for Graph Partitioning in the Presence of NoiseAravindan Vijayaraghavan · Learning, Algorithm Design and Beyond Worst-Case Analysis
• 52mNew Algorithms For The Graph Alignment ProblemMarc Lelarge · Algorithmic Advances for Statistical Inference with Combinatorial Structure
• 43mInformation-Theoretic Bounds and Phase Transitions in Community Detection and High-Dimensional ClustCris Moore · Random Instances and Phase Transitions (Simons Institute, 2016)