Systems and methods for communication system intent analysis
The present disclosure relates generally to systems and methods for analyzing intent. Intents may be analyzed to determine to which device or agent to route a communication. The analyzed intent information can also be used to formulate reports and analyze the accuracy of the identified intents with respect to the received communication.
1. A computer-implemented method comprising:
parsing a communication from an interaction involving a user device to identify one or more operative words from the communication;
determining an intent value selected from a list of intent categories, wherein the intent value is associated with the one or more operative words in the communication, and wherein the intent value and the one or more operative words are associated with an action for the user device associated with the communication;
calculating a metric for the intent value using a projection model, wherein the metric is based on a qualitative strength of an association between the one or more operative words in the communication and the intent value, and wherein the metric is calculated using a plurality of communications associated with the intent value, the interaction, and one or more human agent profiles;
matching the communication with a terminal device associated with a human agent profile, wherein matching is based on the metric;
tracking user feedback and usage data associated with the interaction; and
automatically tracking shifts in user expectations by updating the projection model with the user feedback and usage data, wherein the shifts in user expectations comprise changes or shifts in associations between operative words and intent categories that occur over time in user groups associated with a plurality of users and a plurality of user devices comprising the user device.
2. The computer-implemented method of claim 1 , wherein tracking the user feedback and usage data comprises:
receiving the user feedback and usage data associated with the interaction; and
updating a machine learning algorithm associated with the projection model with updated training using the user feedback and usage data.
3. The computer-implemented method of claim 1 , further comprising:
training an intent management engine using a set of annotation data, wherein the set of annotation data is used by the intent management engine to associate intent values with operative words.
4. The computer-implemented method of claim 1 , further comprising selecting a selected human agent profile of the one or more human agent profiles, wherein the selected human agent profile is selected using a machine learning algorithm trained to identify a correlation of the selected human agent profile to the intent value.
5. The computer-implemented method of claim 1 , wherein tracking the user feedback and usage data associated with the interaction further comprises:
receiving the user feedback and usage data associated with a plurality of interactions including the interaction;
updating a machine learning algorithm using the user feedback and usage data, wherein the machine learning algorithm is used for matching the communication with the terminal device; and
facilitating a new interaction using the machine learning algorithm as updated using the user feedback and usage data.
6. The computer-implemented method of claim 1 , wherein the one or more operative words are identified by comparing words from the communication to one or more known operative words in a database to match words from the communication with the one or more known operative words.
7. The computer-implemented method of claim 1 , wherein determination of the intent value is a probabilistic selection based on training of a machine learning system.
8. The computer-implemented method of claim 1 , wherein the projection model is a neural network comprising neural network settings that can be modified with real-time dynamic feedback from the user feedback and usage data to track probabilistic shifts in the user expectations from the user feedback and usage data with the neural network.
9. The computer-implemented method of claim 1 , wherein the shifts in user expectations comprise changes or shifts in associations between operative words and intent categories that occur over time in a geographic area.
10. A user device comprising:
a memory; and
one or more processors coupled to the memory, the one or more processors configured to perform operations comprising:
parsing a communication from an interaction involving the user device to identify one or more operative words from the communication;
determining an intent value selected from a list of intent categories, wherein the intent value is associated with the one or more operative words in the communication, and wherein the intent value and the one or more operative words are associated with an action for the user device associated with the communication;
calculating a metric for the intent value using a projection model, wherein the metric is based on a qualitative strength of an association between the one or more operative words in the communication and the intent value, and wherein the metric is calculated using a plurality of communications associated with the intent value, the interaction, and one or more human agent profiles;
matching the communication with a terminal device associated with a human agent profile, wherein matching is based on the metric;
tracking user feedback and usage data associated with the interaction; and
automatically tracking shifts in user expectations by updating the projection model with the user feedback and usage data, wherein the shifts in user expectations comprise changes or shifts in associations between operative words and intent categories that occur over time in user groups associated with a plurality of users and a plurality of user devices comprising the user device.
11. The user device of claim 10 , wherein tracking the user feedback and usage data comprises:
receiving the user feedback and usage data associated with the interaction; and
updating a machine learning algorithm associated with the projection model with updated training using the user feedback and usage data.
12. The user device of claim 10 , wherein the one or more processors are configured for operations further comprising:
training an intent management engine using a set of annotation data, wherein the set of annotation data is used by the intent management engine to associate intent values with operative words.
13. The user device of claim 10 , wherein the one or more processors are configured for operations further comprising selecting a selected human agent profile of the one or more human agent profiles, wherein the selected human agent profile is selected using a machine learning algorithm trained to identify a correlation of the selected human agent profile to the intent value.
14. The user device of claim 10 , wherein tracking the user feedback and usage data associated with the interaction further comprises:
receiving the user feedback and usage data associated with a plurality of interactions including the interaction;
updating a machine learning algorithm using the user feedback and usage data, wherein the machine learning algorithm is used for matching the communication with the terminal device; and
facilitating a new interaction using the machine learning algorithm as updated using the user feedback and usage data.
15. The user device of claim 10 , wherein the one or more operative words are identified by comparing words from the communication to one or more known operative words in a database to match words from the communication with the one or more known operative words.
16. The user device of claim 10 , wherein determination of the intent value is a probabilistic selection based on training of a machine learning system.
17. The user device of claim 10 , wherein the projection model is a neural network comprising neural network settings that can be modified with real-time dynamic feedback from the user feedback and usage data to track probabilistic shifts in the user expectations from the user feedback and usage data with the neural network.
18. The user device of claim 10 , wherein the shifts in user expectations comprise changes or shifts in associations between operative words and intent categories that occur over time in a geographic area.
19. A non-transitory computer readable medium comprising instructions that, when executed by one or more processors of a user device, cause the user device to perform operations comprising:
parsing a communication from an interaction involving the user device to identify one or more operative words from the communication;
determining an intent value selected from a list of intent categories, wherein the intent value is associated with the one or more operative words in the communication, and wherein the intent value and the one or more operative words are associated with an action for the user device associated with the communication;
calculating a metric for the intent value using a projection model, wherein the metric is based on a qualitative strength of an association between the one or more operative words in the communication and the intent value, and wherein the metric is calculated using a plurality of communications associated with the intent value, the interaction, and one or more human agent profiles;
matching the communication with a terminal device associated with a human agent profile, wherein matching is based on the metric;
tracking user feedback and usage data associated with the interaction; and
automatically tracking shifts in user expectations by updating the projection model with the user feedback and usage data, wherein the shifts in user expectations comprise changes or shifts in associations between operative words and intent categories that occur over time in user groups associated with a plurality of users and a plurality of user devices comprising the user device.
20. The non-transitory computer readable medium of claim 19 , wherein tracking the user feedback and usage data comprises:
receiving the user feedback and usage data associated with the interaction; and
updating a machine learning algorithm associated with the projection model with updated training using the user feedback and usage data.
21. The non-transitory computer readable medium of claim 19 , wherein the instructions configure the user device for operations further comprising:
training an intent management engine using a set of annotation data, wherein the set of annotation data is used by the intent management engine to associate intent values with operative words.
22. The non-transitory computer readable medium of claim 19 , wherein the instructions configure the user device for operations further comprising selecting a selected human agent profile of the one or more human agent profiles, wherein the selected human agent profile is selected using a machine learning algorithm trained to identify a correlation of the selected human agent profile to the intent value.
23. The non-transitory computer readable medium of claim 19 , wherein tracking the user feedback and usage data associated with the interaction further comprises:
receiving the user feedback and usage data associated with a plurality of interactions including the interaction;
updating a machine learning algorithm using the user feedback and usage data, wherein the machine learning algorithm is used for matching the communication with the terminal device; and
facilitating a new interaction using the machine learning algorithm as updated using the user feedback and usage data.
24. The non-transitory computer readable medium of claim 19 , wherein the one or more operative words are identified by comparing words from the communication to one or more known operative words in a database to match words from the communication with the one or more known operative words.
25. The non-transitory computer readable medium of claim 19 , wherein determination of the intent value is a probabilistic selection based on training of a machine learning system.
26. The non-transitory computer readable medium of claim 19 , wherein the projection model is a neural network comprising neural network settings that can be modified with real-time dynamic feedback from the user feedback and usage data to track probabilistic shifts in the user expectations from the user feedback and usage data with the neural network.
27. The non-transitory computer readable medium of claim 19 , wherein the shifts in user expectations comprise changes or shifts in associations between operative words and intent categories that occur over time in a geographic area.