Customer advocacy through a virtual assistant coupled to a contact center
The technology disclosed herein includes analyzing a transcript of a communications session between a customer operating a customer device and a virtual assistant to determine a customer query; analyzing the customer query to generate supplemental information to respond to the customer query; and providing the supplemental information to the customer device by the virtual assistant.
1. A method implemented on a computing system, wherein the method comprises:
analyzing a transcript, by a service analyzer of a contact center application being executed by the computing system, of a communications session between a customer operating a customer device and a virtual assistant to determine a customer query;
the service analyzer analyzing the customer query and a search engine generating supplemental information from a customer interaction database including extracted information from the transcript and an original conversation between an agent and a customer, the customer interaction database including the supplemental information stored in response to the customer query, wherein analyzing the customer query further comprises:
extracting one or more triples describing relationships between entities and objects of the customer query;
aligning the one or more triples with a seed dictionary including a collection of linguistic rules based in part on customer-based artifact;
extracting one or more topics from the customer query by categorizing text of the customer query and segmenting the customer query into a plurality of discrete bins; and
storing one or more aligned triples and one or more topics in the customer interaction database; and
the contact center application sending the supplemental information during the communication session or in a subsequent communication session to the customer device by the virtual assistant.
2. The method of claim 1 , comprising providing the supplemental information to the customer device in real-time during the communications session.
3. The method of claim 1 , comprising translating the customer query to a correct syntax prior to analyzing the customer query.
4. The method of claim 1 , comprising receiving communications data from the communications session from the virtual assistant via an application programming interface (API), storing the communications data as session data, and converting recorded speech of the session data into text to create the transcript.
5. The method of claim 1 , comprising providing the supplemental information from a service analyzer of the contact center application to the virtual assistant via an application programming interface (API).
6. The method of claim 1 , wherein analyzing the transcript to determine the customer query comprises isolating customer input data from the transcript, determining at least one of an intent and topic of the customer input data, performing semantic tagging of the customer input data, analyzing dialog of the customer input data, and analyzing one or more of the at least one of intent and topic, semantic tagging, and analyzed dialog to determine the customer query.
7. The method of claim 1 , comprising generating the supplemental information by generating a search query from the customer query and the customer interaction database, executing the search query by a search engine from at least one of internal documents and external documents, and filtering results of the search engine.
8. The method of claim 1 , comprising updating the seed dictionary with the one or more triples.
9. At least one non-transitory machine-readable storage medium operable on a computer and comprising instructions that, when executed, cause at least one processor of the computer to:
analyze, using a service analyzer of a contact center application, a transcript of a communications session between a customer operating a customer device in communication with the contact center application and a virtual assistant to determine a customer query;
the service analyzer analyzing the customer query and a search engine generating supplemental information from a customer interaction database including extracted information from the transcript and an original conversation between an agent and a customer, the customer interaction database including the supplemental information in response to the customer query;
the contact center application sending the supplemental information during the communication session or in a subsequent communication session to the customer device by the virtual assistant; and
wherein the instructions when executed further cause the at least one processor to
direct the contact center application, when analyzing the customer query, to extract one or more triples describing relationships between entities and objects of the customer query;
align the one or more triples with a seed dictionary including a collection of linguistic rules;
extract one or more topics from the customer query by categorizing text of the customer query and segmenting the customer query into a plurality of discrete bins; and
store one or more aligned triples and one or more topics in the customer interaction database.
10. The at least one non-transitory machine-readable storage medium of claim 9 comprising instructions that, when executed, cause at least one processor to:
provide the supplemental information to the customer device in real-time during the communications session.
11. The at least one non-transitory machine-readable storage medium of claim 9 , wherein instructions to analyze the transcript to determine the customer query comprise instructions when executed to isolate customer input data from the transcript, determine at least one of an intent and topic of the customer input data, perform semantic tagging of the customer input data, analyze dialog of the customer input data, and analyze one or more of the at least one of intent and topic, semantic tagging, and analyzed dialog to determine the customer query.
12. The at least one non-transitory machine-readable storage medium of claim 9 , wherein instructions to generate the supplemental information comprise instructions when executed to generate a search query from the customer query and the customer interaction database, execute the search query by a search engine from at least one of internal documents and external documents, and filter results of the search engine.
13. The at least one non-transitory machine-readable storage medium of claim 9 , comprising instructions, when executed to provide metrics data to the customer device, the metrics data comprising anticipated wait time.
14. An electronic computing apparatus comprising:
a memory to store instructions; and
a processor, coupled to the memory, configured to execute the instructions to
analyze, using a service analyzer of a contact center application, a transcript of a communications session between a customer operating a customer device and a virtual assistant to determine a customer query;
generate, using the service analyzer, the customer query and a search engine, supplemental information from a customer interaction database including extracted information from the transcript and an original conversation between an agent and a customer, the customer interaction database including supplemental information in response to the customer query;
send, by a contact center application, the supplemental information during the communication session or in a subsequent communication session to the customer device by the virtual assistant; and
the contact center application (a) extracting one or more triples describing relationships between entities and objects of the customer query; (b) aligning the one or more triples with a seed dictionary including a collection of linguistic rules based in part on customer-based artifact; (c) extracting one or more topics from the customer query by categorizing text of the customer query and segmenting the customer query into a plurality of discrete bins; and (d) storing one or more aligned triples and one or more topics in the customer interaction database.
15. The apparatus of claim 14 , wherein instructions when executed to generate the supplemental information comprise instructions when executed to generate a search query from the customer query and the customer interaction database, execute the search query by a search engine from at least one of internal documents and external documents, and filter results of the search engine.
16. The apparatus of claim 15 , wherein the internal documents comprise company-specific documents available over a private intranet.
17. The apparatus of claim 15 , wherein the instructions when executed cause the processor to categorize the supplemental information according to one or more of problem domains, technologies, products, and company departments.