US 12,147,452 B2
Novelty detection system
Ryan Wright, Portland, OR (US)
Assigned to THATDOT, INC., Portland, OR (US)
Filed by THATDOT, INC., Portland, OR (US)
Filed on Jun. 23, 2021, as Appl. No. 17/356,191.
Claims priority of provisional application 63/043,727, filed on Jun. 24, 2020.
Prior Publication US 2021/0406288 A1, Dec. 30, 2021
Int. Cl. G06F 16/28 (2019.01); G06F 16/901 (2019.01); G06F 18/22 (2023.01); G06F 18/2433 (2023.01); G06N 7/01 (2023.01)
CPC G06F 16/287 (2019.01) [G06F 16/288 (2019.01); G06F 16/9027 (2019.01); G06F 18/22 (2023.01); G06F 18/2433 (2023.01); G06N 7/01 (2023.01)] 20 Claims
OG exemplary drawing
 
1. A system, comprising:
at least one processor; and
at least one memory storing instructions that, in response to execution by the at least one processor, cause the system to:
store, in a graph data structure, a representation of an observation, the representation comprising at least a first node indicative of a first piece of categorical information and a first number of times the first piece of categorical information has been observed in a stream of observations, and a second node indicative of a second piece of categorical information and a second number of times the second piece of categorical information has been observed in the stream of observations;
update the first node to indicate that the first piece of categorical information has been observed again in the stream of observations;
calculate, using the graph data structure, a score indicative of a degree to which the observation is an outlier with respect to other observations read from the stream of observations, the score calculated based at least in part on one or more probabilities computed based, at least in part, on the first number of times the first piece of categorical information has been observed in the stream of observations, and the second number of times the second piece of categorical information has been observed in the stream of observations; and
send the score to a client process.