US 11,727,248 B2
Interpretable node embedding
Zhao Xu, Heidelberg (DE); and Giuseppe Serra, Heidelberg (DE)
Assigned to NEC LABORATORIES EUROPE GMBH, Heidelberg (DE)
Filed by NEC Laboratories Europe GmbH, Heidelberg (DE)
Filed on Apr. 7, 2020, as Appl. No. 16/841,762.
Prior Publication US 2021/0319280 A1, Oct. 14, 2021
Int. Cl. G06N 3/04 (2023.01); G06N 3/08 (2023.01)
CPC G06N 3/04 (2013.01) [G06N 3/08 (2013.01)] 16 Claims
OG exemplary drawing
 
1. A method for extracting human-interpretable entity profiles from a text-labeled data graph of a system comprised of entities, the method comprising:
constructing neural network layers, wherein the data graph comprises nodes representing the entities and edges between the nodes representing links between the entities, wherein a plurality of text is respectively associated with the corresponding edges, and wherein the neural network layers are constructed such that each of the edges between a pair of the nodes is modeled as a function of the associated text and cluster representations of the pair of the nodes,
for each one of the pair of nodes, performing machine learning to learn a tensor to capture patterns among the associated text and the pair of nodes, and
extracting the human-interpretable entity profiles from the tensor.