Methods involving computing correlation anomaly scores
First Claim
1. A method for detecting an error operation of physical sensors in a system, the method comprising:
- defining a first similarity matrix for a target run of data, wherein the target run of data includes data obtained from an N number of physical sensors in the system;
defining a second similarity matrix for a reference run of data, wherein the reference run of data includes data obtained from the N number of physical sensors in the system;
developing a k-neighborhood graph Ni of the i-th node for the target run of data, wherein the k-neighborhood graph of the i-th node is defined as a graph comprising the i-th node and its k-nearest neighbors (NN);
developing a k-neighborhood graph Ni of the i-th node for the reference run of data;
defining a probability distribution p(j/i), wherein p(j/i) is the probability that the j-th node becomes one of the k-NN of the i-th node;
summing the probabilities between the i-th node and the neighbors of the i-th node;
determining an anomaly score of the i-th node from the sum of the probabilities;
wherein the anomaly score is defined as max{|ei(Ni)|,|ei( Ni)−
ē
i( Ni)|}, wherein ei(Ni) is the sum of the probabilities of the target data as a function of the target data, ē
i (Ni) is the sum of the probabilities of the reference data as a function of the target data, ei( Ni) is the sum of the probabilities of the target data as a function of the reference data, and ē
i( Ni) is the sum of the probabilities of the reference data as a function of the reference data;
determining whether the target run of data has changed from the reference run of data responsive to determining the anomaly score of the i-th node;
comparing the anomaly score of the i-th node with a threshold value to determine whether the anomaly score of the i-th node exceeds the threshold value responsive to determining that the target run of data has changed from the reference run of data; and
outputting the anomaly score responsive to determining that the anomaly score of the i-th node exceeds the threshold value, wherein the output anomaly score is indicative of the error operation in the physical sensors of the system.
1 Assignment
0 Petitions
Accused Products
Abstract
An exemplary method for computing correlation anomaly scores, including, defining a first similarity matrix for a target run of data, the target run of data includes an N number of sensors, defining a second similarity matrix for a reference run of data, the target run of data includes the N number of sensors, developing a k-neighborhood graph Ni of the i-th node for the target run of data, wherein the k-neighborhood graph of the i-th node is defined as a graph comprising the i-th node and its k-nearest neighbors (NN), developing a k-neighborhood graph
26 Citations
1 Claim
-
1. A method for detecting an error operation of physical sensors in a system, the method comprising:
-
defining a first similarity matrix for a target run of data, wherein the target run of data includes data obtained from an N number of physical sensors in the system; defining a second similarity matrix for a reference run of data, wherein the reference run of data includes data obtained from the N number of physical sensors in the system; developing a k-neighborhood graph Ni of the i-th node for the target run of data, wherein the k-neighborhood graph of the i-th node is defined as a graph comprising the i-th node and its k-nearest neighbors (NN); developing a k-neighborhood graph N i of the i-th node for the reference run of data;defining a probability distribution p(j/i), wherein p(j/i) is the probability that the j-th node becomes one of the k-NN of the i-th node; summing the probabilities between the i-th node and the neighbors of the i-th node; determining an anomaly score of the i-th node from the sum of the probabilities; wherein the anomaly score is defined as max{|ei(Ni)|,|ei( N i)−
ē
i(N i)|}, wherein ei(Ni) is the sum of the probabilities of the target data as a function of the target data, ē
i (Ni) is the sum of the probabilities of the reference data as a function of the target data, ei(N i) is the sum of the probabilities of the target data as a function of the reference data, and ē
i(N i) is the sum of the probabilities of the reference data as a function of the reference data;determining whether the target run of data has changed from the reference run of data responsive to determining the anomaly score of the i-th node; comparing the anomaly score of the i-th node with a threshold value to determine whether the anomaly score of the i-th node exceeds the threshold value responsive to determining that the target run of data has changed from the reference run of data; and outputting the anomaly score responsive to determining that the anomaly score of the i-th node exceeds the threshold value, wherein the output anomaly score is indicative of the error operation in the physical sensors of the system.
-
Specification