FSCNMF: Fusing Structure and Content via Non-negative Matrix Factorization for Embedding Information Networks

Network embedding has been a central topic of information network mining in the last couple of years. Network embedding learns a compact lower dimensional vector representation for each node of the network, and uses this lower dimensional representation for different machine learning and mining applications. The existing methods for network embedding consider mainly the structure of the network. But many real world networks also contain rich textual or other type of content associated with each node, which can help to understand the underlying semantics of the network. It is not straightforward to integrate the content of each node in the state-of-the-art network embedding methods. In this work, we propose a nonnegative matrix factorization based optimization framework, namely FSCNMF which considers both the network structure and the content of the nodes while learning a lower dimensional vector representation of each node in the network. Our approach systematically exploits the consistency of the network structure and the content of the nodes. We further extend the basic FSCNMF to an advanced method, namely FSCNMF++ to capture the higher order proximities in the network. We conduct experiments on real world information networks for different types of machine learning applications such as node clustering, visualization, and multi-class classification. The results show that our method can represent the network significantly better than the state-of-the-art algorithms and improve the performance across all the applications that we consider.