US 12,437,003 B2
Identification of influential nodes in graph datasets using combinatorial optimization formulations
Xiaoyuan Liu, San Jose, CA (US); Ankith Mohan, Blacksburg, VA (US); and Sarvagya Upadhyay, San Jose, CA (US)
Assigned to Fujitsu Limited, Kawasaki (JP)
Filed by Fujitsu Limited, Kawasaki (JP)
Filed on Jan. 18, 2023, as Appl. No. 18/156,330.
Prior Publication US 2024/0241910 A1, Jul. 18, 2024
Int. Cl. G06F 16/906 (2019.01); G06F 16/901 (2019.01); G06Q 50/00 (2024.01)
CPC G06F 16/906 (2019.01) [G06F 16/9024 (2019.01); G06Q 50/01 (2013.01)] 15 Claims
OG exemplary drawing
 
1. A method, comprising:
obtaining a graph dataset that includes a plurality of nodes and a plurality of edges between the plurality of nodes, the plurality of nodes representing buildings and the plurality of edges representing roads in between the buildings;
partitioning the graph dataset into a first node cluster and a second node cluster in which the first node cluster and the second node cluster each include one or more nodes from the graph dataset;
identifying, using a digital annealer, one or more first candidate influential nodes included in the first node cluster and one or more second candidate influential nodes included in the second node cluster, wherein identification of the one or more first candidate influential nodes and the one or more second candidate influential nodes are cast as optimization problems presented as a quadratic unconstrained binary optimization (QUBO) function having a format solvable by the digital annealer, the digital annealer configured to solve the optimization problems to identify the one or more first candidate influential nodes and the one or more second candidate influential nodes, wherein the one or more first candidate influential nodes and the one or more second candidate influential nodes represent the buildings connected to the roads with greater traffic flows;
selecting one or more of the first candidate influential nodes and the second candidate influential nodes as influencer nodes, the selecting including:
aggregating the first candidate influential nodes and the second candidate influential nodes as a set of candidate influential nodes; determining a cumulative influence of the set of candidate influential nodes based on summing respective influences of each node included in the set of candidate influential nodes on each other node included in the graph dataset but not included in the set of candidate influential nodes; and selecting the influencer nodes from the set of candidate influential nodes based on respective influences of the first candidate influential nodes and of the second candidate influential nodes exerted on other nodes included in the graph dataset; and
identifying a respective building included in the graph dataset corresponding to each of the influencer nodes.