US 12,346,913 B2
Method, apparatus and computer program product for predictive graph-based network analysis
Chirag Mittal, Gurgaon (IN); Arjun Rana, Noida (IN); Rajesh Sabapathy, Gurgaon (IN); Erin Conway, Minneapolis, MN (US); Sudeep Vishnumurthy, Eden Prairie, MN (US); and Arushi Parganiha, Bhopal (IN)
Assigned to UnitedHealth Group Incorporated, Minnetonka, MN (US)
Filed by UnitedHealth Group Incorporated, Minnetonka, MN (US)
Filed on Mar. 1, 2021, as Appl. No. 17/249,383.
Prior Publication US 2022/0277315 A1, Sep. 1, 2022
Int. Cl. G06Q 30/018 (2023.01); G06F 16/908 (2019.01); G06F 16/951 (2019.01); G06Q 10/0635 (2023.01); G06Q 10/10 (2023.01); G06Q 30/0201 (2023.01); G06Q 40/08 (2012.01); G06Q 50/00 (2012.01)
CPC G06Q 30/0185 (2013.01) [G06F 16/908 (2019.01); G06F 16/951 (2019.01); G06Q 10/0635 (2013.01); G06Q 10/10 (2013.01); G06Q 30/0201 (2013.01); G06Q 40/08 (2013.01); G06Q 50/01 (2013.01)] 19 Claims
OG exemplary drawing
 
1. A system comprising one or more processors and at least one memory storing processor executable instructions that, when executed by the one or more processors, cause the one or more processors to perform operations comprising:
receiving, from a database and based at least in part on input metadata of at least one input data object of a plurality of input data objects, a scheme data object that comprises scheme metadata;
generating a tabular data object based at least in part on the plurality of input data objects and the scheme metadata;
generating a graph data object based at least in part on the tabular data object, wherein:
(i) the graph data object comprises (a) a plurality of nodes and a plurality of node attributes for the plurality of nodes and (b) a plurality of edges and plurality of edge attributes for the plurality of edges,
(ii) the plurality of node attributes and the plurality of edge attributes represent one or more behavioral utilization aspects of a triangular relationship that is (a) among one or more of a plurality of referring provider nodes, one or more of a plurality of servicing provider nodes, and one or more of a plurality of member nodes, and (b) with respect to a fraud scheme corresponding to the scheme data object,
(iii) the plurality of nodes comprises
(a) the plurality of member nodes,
(b) the plurality of referring provider nodes, and
(c) the plurality of servicing provider nodes,
(iv) each edge of the plurality of edges connects two of the plurality of nodes;
(v) one or more columns of the tabular data object are used to identify one or more of the plurality of nodes; and
(vi) the plurality of edges is based at least in part on a plurality of tabular correlations indicated by the tabular data object;
generating a reduced search space for a second stage of processing the graph data object by, in a first stage of processing the graph data object:
identifying, using a first traversal of the graph data object, one or more member nodes of the plurality of member nodes as one or more irregular member nodes based at least in part on detecting, during the first traversal, that the one or more member nodes comprise a first outlier attribute relative to at least one of the plurality of node attributes or the plurality of edge attributes; and
identifying, using a second traversal of the graph data object, one or more referring provider nodes of the plurality of referring provider nodes as one or more irregular referring provider nodes based at least in part on detecting, during the second traversal, that the one or more referring provider nodes comprise at least a second outlier attribute relative to at least one of the plurality of node attributes or the plurality of edge attributes;
in the second stage of processing the graph data object, selectively processing the graph data object within the reduced search space by:
identifying, using a third traversal of the graph data object, a servicing provider node of the plurality of servicing provider nodes as an irregular servicing provider node based at least in part on (i) the one or more irregular member nodes identified in the first traversal connected to the servicing provider node and (ii) the one or more irregular referring provider nodes identified in the second traversal; and
generating a prediction output user interface configured to depict one or more of the one or more irregular member nodes, the one or more irregular referring provider nodes, or the irregular servicing provider node.