GEMSEC: Graph Embedding with Self Clustering

2018/02/28 by Benedek Rozemberczki, Ryan Davies, Rik Sarkar +1 · 3 citations
Computer Science · #cs.SI

paper · pdf

published as ASONAM 2019

arxiv created 2019/07/25 · arxiv updated 2019/07/26

Abstract

Modern graph embedding procedures can efficiently process graphs with millions of nodes. In this paper, we propose GEMSEC -- a graph embedding algorithm which learns a clustering of the nodes simultaneously with computing their embedding. GEMSEC is a general extension of earlier work in the domain of sequence-based graph embedding. GEMSEC places nodes in an abstract feature space where the vertex features minimize the negative log-likelihood of preserving sampled vertex neighborhoods, and it incorporates known social network properties through a machine learning regularization. We present two new social network datasets and show that by simultaneously considering the embedding and clustering problems with respect to social properties, GEMSEC extracts high-quality clusters competitive with or superior to other community detection algorithms. In experiments, the method is found to be computationally efficient and robust to the choice of hyperparameters.

Cited by

• Regularization and Selection in A Directed Network Model with Nodal Homophily and Nodal Effects
• Burning some myths on privacy properties of social networks against active attacks
• Robust Markov stability for community detection at a scale learned based on the structure
• New exponential law for real networks
• Landmark-Based Node Representations for Shortest Path Distance Approximations in Random Graphs