Community detection in graphs — main approaches.

hard

Answer

$(1) Modularity optimization: Louvain, Leiden (greedy hierarchical) — scale to millions of nodes, standard default.$
$(2) Spectral: eigenvectors of the graph Laplacian \to k-means.$
$(3) Label propagation (fast, non-deterministic).$
$(4) Stochastic Block Models (probabilistic).$
$Modern: GNN-based (DeepWalk / Node2Vec + clustering).$
$Leiden fixes Louvain's resolution-limit + connected-community issues — usually the default now.$

Check yourself — multiple choice

Only k-means
Louvain/Leiden (modularity, hierarchical), spectral (Laplacian), label propagation, SBM, GNN embeddings; Leiden = current default
Same as DBSCAN
Random

Community detection: Louvain/Leiden default; spectral / SBM / GNN alternatives.

#clustering#applications

214 interview questions in this topic.