Network community detection via neural embeddings

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Kojaku, Sadamori, Radicchi, Filippo, Ahn, Yong-Yeol, Fortunato, Santo
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929572459577344
author Kojaku, Sadamori
Radicchi, Filippo
Ahn, Yong-Yeol
Fortunato, Santo
author_facet Kojaku, Sadamori
Radicchi, Filippo
Ahn, Yong-Yeol
Fortunato, Santo
contents Recent advances in machine learning research have produced powerful neural graph embedding methods, which learn useful, low-dimensional vector representations of network data. These neural methods for graph embedding excel in graph machine learning tasks and are now widely adopted. However, how and why these methods work -- particularly how network structure gets encoded in the embedding -- remain largely unexplained. Here, we show that node2vec -- shallow, linear neural network -- encodes communities into separable clusters better than random partitioning down to the information-theoretic detectability limit for the stochastic block models. We show that this is due to the equivalence between the embedding learned by node2vec and the spectral embedding via the eigenvectors of the symmetric normalized Laplacian matrix. Numerical simulations demonstrate that node2vec is capable of learning communities on sparse graphs generated by the stochastic blockmodel, as well as on sparse degree-heterogeneous networks. Our results highlight the features of graph neural networks that enable them to separate communities in embedding space.
format Preprint
id arxiv_https___arxiv_org_abs_2306_13400
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Network community detection via neural embeddings
Kojaku, Sadamori
Radicchi, Filippo
Ahn, Yong-Yeol
Fortunato, Santo
Physics and Society
Social and Information Networks
Recent advances in machine learning research have produced powerful neural graph embedding methods, which learn useful, low-dimensional vector representations of network data. These neural methods for graph embedding excel in graph machine learning tasks and are now widely adopted. However, how and why these methods work -- particularly how network structure gets encoded in the embedding -- remain largely unexplained. Here, we show that node2vec -- shallow, linear neural network -- encodes communities into separable clusters better than random partitioning down to the information-theoretic detectability limit for the stochastic block models. We show that this is due to the equivalence between the embedding learned by node2vec and the spectral embedding via the eigenvectors of the symmetric normalized Laplacian matrix. Numerical simulations demonstrate that node2vec is capable of learning communities on sparse graphs generated by the stochastic blockmodel, as well as on sparse degree-heterogeneous networks. Our results highlight the features of graph neural networks that enable them to separate communities in embedding space.
title Network community detection via neural embeddings
topic Physics and Society
Social and Information Networks
url https://arxiv.org/abs/2306.13400