Publica
Hier finden Sie wissenschaftliche Publikationen aus den FraunhoferInstituten. Node2LV: Squared Lorentzian representations for node proximity
 Institute of Electrical and Electronics Engineers IEEE; IEEE Computer Society: IEEE 37th International Conference on Data Engineering, ICDE 2021. Proceedings : Chania, Greece, 1922 April 2021, virtually Los Alamitos, Calif.: IEEE Computer Society Conference Publishing Services (CPS), 2021 ISBN: 9781728191850 ISBN: 9781728191843 pp.20152020 
 International Conference on Data Engineering (ICDE) <37, 2021, Online> 

 English 
 Conference Paper 
 Fraunhofer Singapore () 
Abstract
Recently, network embedding has attracted extensive research interest. Most existing network embedding models are based on Euclidean spaces. However, Euclidean embedding models cannot effectively capture complex patterns, especially latent hierarchical structures underlying in realworld graphs. Consequently, hyperbolic representation models have been developed to preserve the hierarchical information. Nevertheless, existing hyperbolic models only capture the firstorder proximity between nodes. To this end, we propose a new embedding model, named Node2LV, that learns the hyperbolic representations of nodes using squared Lorentzian distances. This yields three advantages. First, our model can effectively capture hierarchical structures that come from the network topology. Second, compared with the conventional hyperbolic embedding methods that use computationally expensive Riemannian gradients, it can be optimized in a more efficient way. Lastly, different from existing hyperbolic embedding models, Node2LV captures higherorder proximities. Specifically, we represent each node with two hyperbolic embeddings, and make the embeddings of related nodes close to each other. To preserve higherorder node proximity, we use a random walk strategy to generate local neighborhood context. We conduct extensive experiments on four different types of realworld networks. Empirical results demonstrate that Node2LV significantly outperforms various graph embedding baselines.