Stat Cafe - Jesus Arroyo
Spectral Clustering in Networks: Overview and Novel Approaches
- Time: Monday 3/31/2025 from 11:30 AM to 12:30 PM
- Location: BLOC 457
Description
Community detection—the task of clustering the nodes of a graph into meaningful groups—has received considerable attention, partly due to its success in analyzing network structures in complex systems. Spectral clustering, which leverages spectral embeddings to represent nodes, is a widely used approach for this problem. In recent years, significant advancements have been made in both the methodology and theory of spectral methods for community detection. This talk will provide an overview of these developments and introduce novel approaches to enhance spectral clustering techniques, with a focus on addressing problems with degree heterogeneity in network data.
Gallery (with photos by Samantha Williams)
