Most of the existing deep-learning-based network analysis tech- niques focus on the problem of learning low-dimensional node repre- sentations. However, networks can also be seen in the light of edges interlinking pairs of nodes. The broad goal of this paper is to intro- duce a deep-learning framework focused on computing edge-centric network embeddings. We present a novel approach called ECNE, which instead of computing edge embeddings by aggregating node embeddings, computes them directly. ECNE leverages the notion of line graph of a graph coupled with an edge weighting mechanism to preserve the dynamic of the original graph in the line graph. We show that ECNE brings benefits wrt the state-of-the-art
Edge-Centric Network Analysis / Pirro', G.. - (2021). (Intervento presentato al convegno 2021 IEEE/ACM conference on Advances in Social Network Analysis and Mining (ASONAM) tenutosi a Virtuale).
Edge-Centric Network Analysis
G. Pirro'
2021
Abstract
Most of the existing deep-learning-based network analysis tech- niques focus on the problem of learning low-dimensional node repre- sentations. However, networks can also be seen in the light of edges interlinking pairs of nodes. The broad goal of this paper is to intro- duce a deep-learning framework focused on computing edge-centric network embeddings. We present a novel approach called ECNE, which instead of computing edge embeddings by aggregating node embeddings, computes them directly. ECNE leverages the notion of line graph of a graph coupled with an edge weighting mechanism to preserve the dynamic of the original graph in the line graph. We show that ECNE brings benefits wrt the state-of-the-artI documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.