| CPC G06F 40/30 (2020.01) [G06F 40/166 (2020.01); G06F 40/284 (2020.01); G06N 7/01 (2023.01); G06N 20/00 (2019.01)] | 20 Claims |
|
1. A computer-implemented method comprising:
in a batch processing mode:
receiving, by a computer, observation data of a plurality of wireless signals observed by a plurality of observer devices;
training, by the computer, a first inductive classifier to generate a first set of deductive rules for syntactic matches between service set identifiers of the plurality of wireless signals and entities associated with the service set identifiers;
applying a multi-dimensional clustering process to generate one or more hyperclusters based on both temporal persistence and spatial proximity of the plurality of wireless signals;
training, by the computer, a second inductive classifier to generate a second set of deductive rules based on the one or more generated hyperclusters, wherein the second inductive classifier is trained to assign a syntactic and/or semantic meaning to a given wireless signal based on a hypercluster of the given wireless signal;
aggregating the first set of deductive rules and the second set of deductive rules to generate an aggregated set of deductive rules for assigning a syntactic and/or semantic meaning to a given wireless signal;
in a real-time deduction mode:
receiving, by the computer, information of an unresolved wireless signal;
generating, by the computer, a semantic resolution of the wireless signal based upon applying one or more rules from the aggregated set of deductive rules; and
displaying, by the computer, the semantic resolution of the wireless signal on a graphical user interface.
|