Intelligent conversational gateway
A method comprises receiving at least one natural language input, and determining an intent of the at least one natural language input. In the method, a virtual assistant of a plurality of virtual assistants is recommended to respond to the at least one natural language input based at least in part on the determined intent, and the at least one natural language input is transmitted to the recommended virtual assistant. The determining and recommending are performed using one or more machine learning models, and the plurality of virtual assistants respectively correspond to a plurality of different functions of an enterprise.
1. A method, comprising:
receiving at least one natural language input;
determining an intent of the at least one natural language input;
recommending a virtual assistant of a plurality of virtual assistants to respond to the at least one natural language input based at least in part on the determined intent;
generating at least one of a natural language word and a natural language phrase to add to the at least one natural language input based, at least in part, on one or more matched tokens from the at least one natural language input;
determining a sentiment of the at least one natural language input, wherein the at least one of the generated natural language word and the generated natural language phrase is based at least in part on the determined sentiment; and
transmitting the at least one natural language input augmented with the at least one of the generated natural language word and the generated natural language phrase to the recommended virtual assistant;
wherein at least the determining of the intent and the recommending are performed using one or more machine learning models;
wherein the plurality of virtual assistants respectively correspond to a plurality of different functions of an enterprise; and
wherein the steps of the method are executed by a processing device operatively coupled to a memory.
2. The method of claim 1 , further comprising identifying a function of the plurality of different functions corresponding to the determined intent.
3. The method claim 2 , further comprising:
determining an additional intent of the at least one natural language input;
recommending an additional virtual assistant of the plurality of virtual assistants to respond to the at least one natural language input based at least in part on the determined additional intent; and
transmitting the at least one natural language input to the recommended additional virtual assistant.
4. The method of claim 3 , further comprising identifying an additional function of the plurality of different functions corresponding to the determined additional intent.
5. The method of claim 1 , further comprising training the one or more machine learning models with data comprising a plurality of natural language statements and a plurality of intents corresponding to the plurality of natural language statements.
6. The method of claim 1 , wherein the one or more machine learning models comprises a bi-directional recurrent neural network with long short-term memory for natural language understanding.
7. The method of claim 1 ,
wherein the determining of the sentiment is performed using the one or more machine learning models; and
wherein the recommending of the virtual assistant is further based on the determined sentiment.
8. The method of claim 1 , wherein the sentiment is one of a negative sentiment, a positive sentiment, a neutral sentiment and an anxious sentiment.
9. The method of claim 7 , further comprising:
determining a priority of the at least one natural language input based on the determined sentiment;
wherein the recommending of the virtual assistant is further based on the determined priority.
10. The method of claim 9 , further comprising training the one or more machine learning models with data comprising a plurality of intents, a plurality of sentiments, a plurality of priorities and respective ones of the plurality of virtual assistants corresponding to the pluralities of intents, sentiments and priorities.
11. The method of claim 1 , wherein:
the determining of the sentiment is performed using the one or more machine learning models; and
the method further comprises training the one or more machine learning models with data comprising a plurality of natural language statements and a plurality of sentiments corresponding to the plurality of natural language statements.
12. The method of claim 1 ,
wherein the generating is performed using the one or more machine learning models.
13. The method of claim 12 , further comprising training the one or more machine learning models with data comprising a plurality of predictors and a plurality of labels corresponding to the plurality of predictors.
14. An apparatus comprising:
a processing device operatively coupled to a memory and configured to:
receive at least one natural language input;
determine an intent of the at least one natural language input;
recommend a virtual assistant of a plurality of virtual assistants to respond to the at least one natural language input based at least in part on the determined intent;
generate at least one of a natural language word and a natural language phrase to add to the at least one natural language input based, at least in part, on one or more matched tokens from the at least one natural language input;
determine a sentiment of the at least one natural language input, wherein the at least one of the generated natural language word and the generated natural language phrase is based at least in part on the determined sentiment; and
transmit the at least one natural language input augmented with the at least one of the generated natural language word and the generated natural language phrase to the recommended virtual assistant;
wherein the processing device is configured to use one or more machine learning models to perform at least the determining of the intent and the recommending; and
wherein the plurality of virtual assistants respectively correspond to a plurality of different functions of an enterprise.
15. The apparatus of claim 14 , wherein the processing device is further configured to train the one or more machine learning models with data comprising a plurality of natural language statements and a plurality of intents corresponding to the plurality of natural language statements.
16. The apparatus of claim 14 , wherein the processing device is further configured to:
use the one or more machine learning models to perform the determining of the sentiment;
wherein the recommending of the virtual assistant is further based on the determined sentiment.
17. The apparatus of claim 14 , wherein the processing device is further configured to use the one or more machine learning models to perform the generating.
18. An article of manufacture comprising a non-transitory processor-readable storage medium having stored therein program code of one or more software programs, wherein the program code when executed by at least one processing device causes said at least one processing device to perform the steps of:
receiving at least one natural language input;
determining an intent of the at least one natural language input;
recommending a virtual assistant of a plurality of virtual assistants to respond to the at least one natural language input based at least in part on the determined intent;
generating at least one of a natural language word and a natural language phrase to add to the at least one natural language input based, at least in part, on one or more matched tokens from the at least one natural language input;
determining a sentiment of the at least one natural language input, wherein the at least one of the generated natural language word and the generated natural language phrase is based at least in part on the determined sentiment; and
transmitting the at least one natural language input augmented with the at least one of the generated natural language word and the generated natural language phrase to the recommended virtual assistant;
wherein at least the determining of the intent and the recommending are performed using one or more machine learning models; and
wherein the plurality of virtual assistants respectively correspond to a plurality of different functions of an enterprise.
19. The article of manufacture of claim 18 ,
wherein the determining of the sentiment is performed using the one or more machine learning models; and
wherein the recommending of the virtual assistant is further based on the determined sentiment.
20. The article of manufacture of claim 19 , wherein:
the program code further causes said at least one processing device to perform the step of determining a priority of the at least one natural language input based on the determined sentiment; and
wherein the recommending of the virtual assistant is further based on the determined priority.