Automatic Dimension Selection for a Non-negative Factorization Approach to Clustering Multiple Random Graphs

2014/06/24 by Nam H. Lee, I-Jeng Wang, Lee, Nam H. +8
Computer Science · Engineering · Mathematics · #Advanced Clustering Algorithms Research #FOS: Computer and information sciences #Face and Expression Recognition #Machine Learning (stat.ML) #Remote-Sensing Image Classification #stat.ML

paper · pdf · doi:10.48550/arxiv.1406.6315

This paper has been withdrawn by the author due to a newer version with overlapping contents

openalex publication_date 2014/06/24 · arxiv created 2014/09/09 · arxiv updated 2014/09/10 · openalex created_date 2016/06/24 · openalex updated_date 2026/07/28

Abstract

We consider a problem of grouping multiple graphs into several clusters using singular value thesholding and non-negative factorization. We derive a model selection information criterion to estimate the number of clusters. We demonstrate our approach using "Swimmer data set" as well as simulated data set, and compare its performance with two standard clustering algorithms.

Citations

• Stochastic blockmodel approximation of a graphon: Theory and consistent estimation

Related

• Techniques for clustering interaction data as a collection of graphs
• SMLSOM: The shrinking maximum likelihood self-organizing map
• Simultaneous Dimensionality and Complexity Model Selection for Spectral Graph Clustering
• Latent Random Steps as Relaxations of Max-Cut, Min-Cut, and More
• Dimensionality Reduction for k-means Clustering