US 12,242,557 B1
Systems for determining anomalous interactions based on characteristic types
Vaibhav Gorde, Seattle, WA (US); Sarthak Ghosh, Seattle, WA (US); Yan Zhang, Bellevue, WA (US); Radu Drossu, Seattle, WA (US); and Asaph Engel, San Diego, CA (US)
Assigned to AMAZON TECHNOLOGIES, INC., Seattle, WA (US)
Filed by AMAZON TECHNOLOGIES, INC., Seattle, WA (US)
Filed on Dec. 15, 2022, as Appl. No. 18/066,854.
Int. Cl. G06F 16/9537 (2019.01); G06F 16/951 (2019.01); G06F 16/9535 (2019.01); G06F 16/9536 (2019.01); G06F 16/9538 (2019.01)
CPC G06F 16/9537 (2019.01) [G06F 16/951 (2019.01); G06F 16/9535 (2019.01); G06F 16/9536 (2019.01); G06F 16/9538 (2019.01)] 20 Claims
OG exemplary drawing
 
1. A system comprising:
one or more non-transitory memories storing computer-executable instructions; and
one or more hardware processors to execute the computer-executable instructions to:
determine a first plurality of interactions, wherein each interaction of the first plurality of interactions is associated with a first corresponding set of characteristics, and wherein each characteristic of the first corresponding set of characteristics includes a respective first characteristic type and a respective first characteristic value;
determine, based on the first plurality of interactions, a first subset of the first plurality of interactions associated with one or more first characteristic values that occur at a first frequency that is greater than a first threshold frequency;
based on the one or more first characteristic values occurring at the first frequency that is greater than the first threshold frequency, determine a first set of characteristic types associated with the one or more first characteristic values of the first subset of the first plurality of interactions;
determine a second plurality of interactions, wherein each interaction of the second plurality of interactions is associated with a second corresponding set of characteristics, and wherein each characteristic of the second corresponding set of characteristics includes a respective second characteristic type and a respective second characteristic value;
determine, for a second subset of the second plurality of interactions, one or more second characteristic values that occur at a second frequency greater than one or more of the first threshold frequency or a second threshold frequency and that correspond to the first set of characteristic types;
determine that output has not been determined using the first set of characteristic types within at least a threshold length of time; and
based on the one or more second characteristic values occurring at the second frequency greater than the one or more of the first threshold frequency or the second threshold frequency, and based on determining that output has not been determined using the first set of characteristic types within at least the threshold length of time, generate an output based on the second subset of the second plurality of interactions.