Alert generator for adaptive closed loop communication system
An alert generator in a communication system for processing a call includes at least one machine learning model generating call classifiers from outputs of an audio signal processor and a natural language processor configure to operate on the call. Heuristic logic is configured to transform the call classifiers into a plurality of weighted sub-metrics for the call, and aggregate normalized Gaussian logic is configured to transform the weighted sub-metrics into a metric control. A threshold analyzer is configured to generate an alert signal to the communication system based on the metric control meeting a condition.
1. An alert generator in a communication system for processing a call, the alert generator comprising:
at least one machine learning model generating call classifiers from outputs of an audio signal processor and a natural language processor configured to operate on the call;
heuristic logic configured to transform the call classifiers into a plurality of weighted sub-metrics for the call;
aggregate normalized Gaussian logic to transform the weighted sub-metrics into a metric control; and
a threshold analyzer configured to generate an alert signal to the communication system based on the metric control meeting a condition.
2. The alert generator of claim 1 , further comprising:
an anomaly detector configured to identify anomalous calls.
3. The alert generator of claim 1 , wherein the alert signal configures the communication system for priority response to the condition.
4. The alert generator of claim 1 , further comprising:
logic to associate with the alert signal portions of the call comprising content that contributed to activation of the alert signal.
5. The alert generator of claim 4 , further comprising a learning function utilizing a call history and one or more of the weighted sub-metrics and the metric control.
6. The alert generator of claim 1 , wherein the call is an active call.
7. The alert generator of claim 6 , wherein the at least one machine learning model comprises an ensemble machine learning model.
8. An alert generation method in a communication system for processing a call, the method comprising:
operating at least one machine learning model on outputs of an audio signal processor and a natural language processor to generate call classifiers;
operating heuristic logic to transform the call classifiers into a plurality of weighted sub-metrics for the call;
applying an aggregate normalized Gaussian transform to convert the weighted sub-metrics into a metric control; and
operating a threshold analyzer to generate an alert signal to the communication system based on the metric control meeting a condition.
9. The method of claim 8 , further comprising:
operating an anomaly detector to identify anomalous calls.
10. The method of claim 8 , wherein the alert signal configures the communication system for priority response to the condition.
11. The method of claim 8 , further comprising:
associating with the alert signal portions of the call comprising content that contributed to activation of the alert signal.
12. The method of claim 8 , wherein the call is an active call.
13. The method of claim 12 , wherein the at least one machine learning model comprises an ensemble machine learning model.
14. The method of claim 8 , further comprising:
applying a learning function utilizing a call history and one or more of the weighted sub-metrics and the metric control to the alert generator.
15. A computing apparatus comprising:
a processor; and
a memory storing instructions that, when executed by the processor, configure the apparatus to:
operate at least one machine learning model on outputs of an audio signal processor and a natural language processor to generate call classifiers;
operate heuristic logic to transform the call classifiers into a plurality of weighted sub-metrics for the call;
apply an aggregate normalized Gaussian transform to convert the weighted sub-metrics into a metric control; and
operate a threshold analyzer to generate an alert signal to the communication system based on the metric control meeting a condition.
16. The computing apparatus of claim 15 , wherein the instructions further configure the apparatus to:
operate an anomaly detector to identify anomalous calls.
17. The computing apparatus of claim 15 , wherein the alert signal configures the communication system for priority response to the condition.
18. The computing apparatus of claim 15 , wherein the instructions further configure the apparatus to:
associate with the alert signal portions of the call comprising content that contributed to activation of the alert signal.
19. The computing apparatus of claim 15 , wherein the instructions further configure the apparatus to:
apply a learning function utilizing a call history and one or more of the weighted sub-metrics and the metric control to the alert generator.
20. The computing apparatus of claim 15 , wherein the at least one machine learning model comprises an ensemble machine learning model.