Device interface matching using an artificial neural network
AssigneeHewlett Packard Enterprise Development LP
InventorsCorey Morrison, Jingya Xu
Life of this patent
Estimating the term…
Abstract
An example method includes receiving first information from a first interface of a network infrastructure device. The method receives second information from a second interface of the network infrastructure device. The method compares a first value to a corresponding second value using a comparison selected, in part, due to a data type of the first value and the second value. The method determines whether the first interface and the second interface have provided matching data, based on inputting an output of the comparison into a first node of an ANN.
Claims19
A method, comprising:
- receiving, at a test device, first information from a first interface of a network infrastructure device;
- receiving, at the test device, second information from a second interface of the network infrastructure device, wherein the first interface and the second interface are retrieving the same information from the network infrastructure device;
- comparing a first value of a first set of values of the first information to a corresponding second value of a second set of values of the second information using a comparison selected, in part, due to a data type of the first value and the second value;
- determining whether the first interface and the second interface have provided matching data, based in part on inputting an output of the comparison of the first value and the second value into a first node of an artificial neural network (ANN);
- comparing, based on a determination that the first interface and the second interface have not provided matching data, an input weight of the first node of the ANN to a significance threshold;
- and
- appending, based on a determination that the input weight of the first node exceeds the significance threshold, information corresponding to the comparison of the first value to the second value to a significant deviations list, wherein the significant deviations list represents possible device defects to be investigated by a device developer.
A test device, comprising:
- processing circuitry;
- and
- a memory including instructions that, when executed by the processing circuitry, cause the test device to:
- receive first information from a first interface of a network infrastructure device;
- receive second information from a second interface of the network infrastructure device;
- wherein the first interface and the second interface are retrieving the same information from the network infrastructure device;
- compare a first set of values of the first information to a corresponding second set of values of the second information using a comparison selected, in part, due to a data type of each respective value of the first set and value of the second set;
- determine whether the first interface and the second interface have provided matching data, based in part on inputting outputs of the comparisons of the first set and the second set into an input layer of nodes of an artificial neural network (ANN);
- compare, based on a determination that the first interface and the second interface have not provided matching data, an input weight of each of the input layer of nodes of the ANN to a significance threshold;
- append, based on a determination that an input weight of a first node of the input layer of nodes exceeds the significance threshold, information corresponding to a comparison of a first value to a second value associated with the first node to a significant deviations list, wherein the significant deviations list represents possible device defects to be investigated by a device developer.
A non-transitory, computer readable medium, comprising instructions that, when executed, cause a device to:
- receive first information from a first interface of a network infrastructure device;
- receive second information from a second interface of the network infrastructure device, wherein the first interface and the second interface are retrieving the same information from the network infrastructure device;
- compare a first set of values of the first information to a corresponding second set of values of the second information, wherein a string datatype values subset of the first set of values are compared to a string datatype values subset of the second set of values using fuzzy string match comparisons;
- determine whether the first interface and the second interface have provided matching data, based in part on inputting outputs of the comparisons of the first set and the second set into an input layer of nodes of an artificial neural network (ANN), wherein outputs of the fuzzy string match comparisons are compared to a match threshold;
- compare, based on a determination that the first interface and the second interface have not provided matching data, an input weight of each of the input layer of nodes of the ANN to a significance threshold;
- append, based on a determination that an input weight of an input node of the input layer of nodes exceeds the significance threshold, information corresponding to a comparison of a first value to a second value associated with the input node to a significant deviations list, wherein the significant deviations list represents possible device defects to be investigated by a device developer.
Dependencies are inferred from the claim text.