Skip to main content
. 2020 Jun 2;18:1414–1428. doi: 10.1016/j.csbj.2020.05.017

Fig. 4.

Fig. 4

Pipeline for representing knowledge graphs in a low dimensional space. Starting with a knowledge graph, this space can be generated using one of the following options: Matrix Factorization (a), Translational Models (b) or Neural Network Models (c). The output of this pipeline is an embedding space that clusters similar node types together.