US 12,147,986 B1
Machine learning creation and usage of cluster visualizations
Paola Gittino, Riverview, FL (US); William George Cormier, Clearwater, FL (US); Steve Thomas Dayton, Brandon, FL (US); and Miriam Silver, Tel Aviv (IL)
Assigned to Citigroup Technology, Inc., Weehawken, NJ (US)
Filed by Citigroup Technology, Inc., Weehawken, NJ (US)
Filed on Oct. 12, 2023, as Appl. No. 18/379,405.
Application 18/379,405 is a continuation in part of application No. 14/707,852, filed on May 8, 2015, abandoned.
Claims priority of provisional application 62/079,110, filed on Nov. 13, 2014.
Int. Cl. G06Q 20/40 (2012.01); G06T 11/20 (2006.01)
CPC G06Q 20/4016 (2013.01) [G06T 11/206 (2013.01); G06T 2200/24 (2013.01)] 18 Claims
OG exemplary drawing
 
1. A method, comprising:
by one or more processors coupled to memory:
automatically generating, using a machine learning algorithm, a network model of each interacting user of a plurality of interacting users of an institution, the network model being based on variables of interactions of each user comprising each of density, size, average degree, average path length, and a clustering coefficient;
automatically generating, using the machine learning algorithm, a visualization in a graphical user interface of encoded data in the network model comprising encoded data attributes of each of the plurality of interacting users relating to user risk level, user history of reports to authorities, user type, and user interactions for interacting users having both direct and indirect connections to one another and interacting users both with and without histories of anomalous interactions displayed as visual nodes;
inputting into the machine learning algorithm, network analysis variables to obtain a level of local clustering based on the network analysis variables, wherein the network analysis variables comprise a ratio of links connecting a plurality of pairs to at least one member of each of a plurality of communities of icons representing an interacting user having a history of anomalous interactions to a maximum possible number of links connecting between the plurality of pairs of icons, wherein the machine learning algorithm is trained to determine the level of local clustering based on patterns, connections, correlations, or trends within the network analysis variables;
automatically detecting, using the machine learning algorithm, at least one community of the plurality of communities of icons within the plurality of pairs of icons having or exceeding a predetermined level of local clustering indicative of a suspicious activity based on the level of local clustering; and
automatically generating, using the machine learning algorithm, the visualization of a size of the at least one community of the plurality of communities of icons based on a level of associated risk determined based on the level of local clustering.