it's an incomplete graph problem, no all the edges (and vertices) are available, i guess the best way to describe it is finding missing edges by identifying network "hubs" i.e. most influential nodes (as in PageRank or Katz Centrality https://en.wikipedia.org/wiki/Katz_centrality ). The original paper: https://arxiv.org/abs/2109.07181v2

It's not a database problem but more of a graph search algorithm optimized for particular network topology.