| CPC G06F 16/9535 (2019.01) [G06F 9/547 (2013.01); G06F 16/9024 (2019.01); G06F 16/9538 (2019.01); G06Q 30/0631 (2013.01)] | 6 Claims |
|
1. One or more non-transitory computer-readable media storing instructions which, when executed by one or more processors, cause:
receiving, at a server computer, a representation of a network graph comprising a plurality of nodes and a plurality of edges,
the plurality of nodes comprising a plurality of creator nodes and a plurality of content nodes,
each content node corresponding to a respective content type of a plurality of content types defined by an ontology,
each creator node corresponding to a respective content creator,
each creator node being associated with one or more respective content types and a respective set of followers, and
each creator node being connected, by an edge, to each content node that corresponds to a content type with which that creator node is associated;
programmatically determining, using a community detection algorithm, a plurality of communities detected to exist within the network graph, each community comprising both a corresponding set of community nodes which is a subset of the plurality of nodes of the network graph and a corresponding set of community edges that connect certain nodes of the corresponding set of community nodes based on the community detection algorithm;
receiving, at the server computer, an input comprising a specification of a particular content creator;
programmatically identifying the creator node corresponding to the particular content creator and one or more particular content types with which that creator node is most strongly associated;
programmatically identifying one or more particular communities that comprise a content node corresponding to one of the one or more particular content types;
programmatically identifying a set of specific creator nodes associated with at least one of the one or more particular content types and comprised by at least one of the one or more particular communities;
for each node of the set of specific creator nodes, programmatically determining one or more specific content types with which that creator node is most strongly associated;
for each node of the set of specific creator nodes, programmatically calculating a respective specific centrality measure of that node between each content node corresponding to each of the particular content types and each content node corresponding to each of the specific content types, based, in part, on stored characterizations of the set of followers of each node of the set of specific creator nodes;
programmatically generating recommendation data identifying a prime content creator corresponding to the specific creator node with the highest calculated specific centrality measure, the recommendation data also identifying the content type most strongly associated with the specific creator node of the prime content creator; and
causing to be displayed, in a graphical user interface displayed on a display of a client computing device, the recommendation data.
|