Machine-learning digital assistants

Abstract

A method of improving response times associated with responding to requests submitted at one or more front-end systems is disclosed. An utterance is listened for at an intelligent virtual assistant included in the one or more front-end systems. At least one of an intent, a context, and a classification is inferred from the utterance. One or more back-end system commands are generated based on the inferring. The one or more back-end system commands are selected based on machine-learned mappings of the at least one of the intent, the context, and the classification to machine-learned organization-specific pathways into the one or more back-end systems. The one or more back-end system commands are distributed across the one or more back-end systems. A response to the utterance is communicated for presentation via the intelligent virtual assistant, the response including an aggregation of the one or more results received.

1 . (canceled) 
     
     
         2 . A system comprising:
 one or more computer processors;   one or more computer memories;   a set of instructions stored in the one or more computer memories, the set of instructions configuring the one or more computer processors to perform operations, the operations comprising:   listening for an utterance at an intelligent virtual assistant included in one or more front-end systems, the utterance including a voice utterance or a chat utterance;   inferring an intent from the utterance;   generating a request for a task to be performed at a node, the generating of the request based on the inferring of the intent;   generating a plurality of back-end system commands for performing the task based on a convergence of machine-learned pathways between the node and the back-end systems, the convergence identified based on an application of a first machine-learned algorithm and an application of a second machine-learned algorithm, the first machine-learned algorithm identifying a first set of the pathways going from the node into the one or more back-end systems, the second machine-learned algorithm identifying pathways going from the one or more back-end systems to the node; and   generating a response to the utterance for presentation via the intelligent virtual assistant, the response including an aggregation of the one or more results.   
     
     
         3 . The system of  claim 2 , wherein the vocabulary is machine-learned from past utterances received in an environment and past distributions of back-end system commands in the environment. 
     
     
         4 . The system of  claim 3 , wherein a training data set corresponding the machine-learned vocabulary includes feature vectors specifying intents corresponding to the past utterances. 
     
     
         5 . The system of  claim 2 , wherein the vocabulary is an insight that is derived from a comparison of the environment to other environments having an attribute in common with the environment, the attribute pertaining to at least one of an industry type associated with the environment or a size of the environment, 
     
     
         6 . The system of  claim 2 , wherein be one or more back-end systems include a private data source and a public data source, the private data source being private to an environment, the private data source being correlated with the public data source based on HTML-based patterns reflecting structures of the private data source and the public data source. 
     
     
         7 . The system of  claim 2 , further comprising formatting the response based on machine-learned rules associated with an environment, the machine-learned rules specifying a graphical layout, for the response based on the intent. 
     
     
         8 . The system of  claim 2 , further comprising:
 deriving insights pertaining to an environment, the insights identifying a subset of the machine-learned pathways that are specific to one or more attributes of the environment; and   storing the insights in a cloud-based database for anonymized access in one or more additional environments.   
     
     
         9 . A method comprising:
 listening for an utterance at an intelligent virtual assistant included in one or more front-end systems, the utterance including a voice utterance or a chat utterance;   inferring an intent from the utterance;   generating a request for a task to be performed at a node, the generating of the request based on the inferring of the intent;   generating a plurality of back-end system commands for performing the task based on a convergence of machine-learned pathways between the node and the back-end systems, the convergence identified based on an application of a first machine-learned algorithm and an application of a second machine-learned algorithm, the first machine-learned algorithm identifying a first set of the pathways going from the node into the one or more back-end systems, the second machine-learned algorithm identifying pathways going from the one or more back-end systems to the node; and   generating a response to the utterance for presentation via the intelligent virtual assistant, the response including an aggregation of the one or more results.   
     
     
         10 . The method of  claim 9 , wherein the vocabulary is machine-learned from past utterances received in an environment and past distributions of back-end system commands in the environment. 
     
     
         11 . The method of  claim 10 , wherein a training data set corresponding to the machine-learned vocabulary includes feature vectors specifying intents corresponding to the past utterances. 
     
     
         12 . The method of  claim 9 , wherein the vocabulary is an insight that is derived from a comparison of the environment to other environments having an attribute in common with the environment, the attribute pertaining to at least one of an industry type associated with the environment or a size of the environment. 
     
     
         13 . The method of  claim 9 , wherein the one or more back-end systems include a private data source and a public data source, the private data source being private to an environment, the private data source being correlated with the public data source based on HTML-based patterns reflecting structures of the private data source and the public data source. 
     
     
         14 . The method of  claim 9 , further comprising formatting the response based on machine-learned rules associated with an environment, the machine-learned rules specifying a graphical layout for the response based on the intent. 
     
     
         15 . The method of  claim 9 , further comprising:
 deriving insights pertaining to an environment, the insights identifying a subset of the machine-learned pathways that are specific to one or more attributes of the environment; and   storing the insights in a cloud-based database for anonymized access in one or more additional environments.   
     
     
         16 . A non-transitory computer-readable storage medium storing instructions thereon, which, when executed by one or more processors, cause the one or more processors to perform operations, the operations comprising:
 listening for an utterance at an intelligent virtual assistant included in one or more front-end systems, the utterance including a voice utterance or a chat utterance;   inferring an intent from the utterance;   generating a request for a task to be performed at a node, the generating of the request based on the inferring of the intent;   generating a plurality of back-end system commands for performing the task based on a convergence of machine-learned pathways between the node and the back-end systems, the convergence identified based on an application of a first machine-learned algorithm and an application of a second machine-learned algorithm, the first machine-learned algorithm identifying a first set of the pathways going from the node into the one or more back-end systems, the second machine-learned algorithm identifying pathways going from the one or more back-end systems to the node; and   generating a response to the utterance for presentation via the intelligent virtual assistant, the response including an aggregation of the one or more results.   
     
     
         17 . The non-transitory computer-readable storage medium of  claim 16 , wherein the vocabulary is machine-learned from past utterances received in an environment and past distributions of hack-end system commands in the environment. 
     
     
         18 . The non-transitory computer-readable storage medium of  claim 17 , wherein a training data set corresponding to the machine-learned vocabulary includes feature vectors specifying intents corresponding to the past utterances. 
     
     
         19 . The non-transitory computer-readable storage medium of  claim 16 , wherein the vocabulary is an insight that is derived from a comparison of the environment to other environments having an attribute in common with the environment, the attribute pertaining to at least one of an industry type associated with the environment or a size of the environment. 
     
     
         20 . The non-transitory computer-readable storage medium of  claim 16 , wherein the one or more hack-end systems include a private data source and a public data source, the private data source being private to an environment, the private data source being correlated with the public data source based on HTML-based patterns reflecting structures of the private data source and the public data source. 
     
     
         21 . The non-transitory computer-readable storage medium of  claim 16 , further comprising formatting the response based on machine-learned rules associated with an environment, the machine-learned rules specifying a graphical layout for the response based on the intent.