SIMULTANEOUS ATTRIBUTED NETWORK EMBEDDING AND CLUSTERING

Lazhar labiod, Mohamed Nadif

Keywords: Attributed network, Embedding, clustering, matrix decomposition, spectral rotation

TL;DR: This paper propose a novel matrix decomposition framework for simultaneous attributed network data embedding and clustering.

Abstract: To deal simultaneously with both, the attributed network embedding and clustering, we propose a new model. It exploits both content and structure information, capitalising on their simultaneous use. The proposed model relies on the approximation of the relaxed continuous embedding solution by the true discrete clustering one. Thereby, we show that incorporating an embedding representation provides simpler and more interpretable solutions. Experiment results demonstrate that the proposed algorithm performs better, in terms of clustering and embedding, than the state-of-art algorithms, including deep learning methods devoted to similar tasks for attributed network datasets with different proprieties.