Example-driven machine learning scheme for dialog system engines

Abstract

A method for example-driven machine learning is disclosed herein. The method comprises maintaining a plurality of dialog system rules and a knowledge database including a plurality of intent objects and a plurality of entity objects. The plurality of intent objects and the plurality of entity objects are associated with at least one dialog system rule. An exemplary phrase is received and one or more linguistic elements are retrieved from the exemplary phrase. It is determined that at least one of the linguistic elements is directed to at least one of the plurality of intent objects of the plurality of entity objects and the at least one of the linguistic elements in association with the at least one dialog system rule is added to the knowledge database.

What is claimed is: 
     
         1 . A method, comprising:
 maintaining, by a server comprising at least one processor and a memory storing processor-executable codes, a plurality of dialog system rules;   maintaining, by the server, a knowledge database, wherein the knowledge database includes a plurality of intent objects and a plurality of entity objects, wherein the plurality of intent objects and the plurality of entity objects are associated with at least one dialog system rule;   receiving, by the server, an exemplary phrase;   retrieving, by the server, one or more linguistic elements from the exemplary phrase;   determining, by the server, that at least one of the linguistic elements is directed to at least one of the plurality of intent objects or the plurality of entity objects; and   adding, by the server, the at least one of the linguistic elements in association with the at least one dialog system rule to the knowledge database.   
     
     
         2 . The method of  claim 1 , further comprising:
 providing, via a developer interface, the at least one of the linguistic elements for review, wherein the providing includes displaying the plurality of dialog system rules and one or more elements related with the plurality of dialog system rules;   receiving, via the developer interface, one or more corrections to the at least one of the linguistic elements, wherein the one or more corrections include associating the at least one of the linguistic elements with the one or more elements of the plurality of dialog system rules based on user selections; and   updating, by the server, the at least one of the linguistic elements based on the one or more corrections.   
     
     
         3 . The method of  claim 1 , further comprising:
 receiving, by a dialog system interface, a user request from a user;   pre-processing, by the dialog system interface, the user request; and   transmitting, by the dialog system interface, the user request to the server.   
     
     
         4 . The method of  claim 3 , wherein the pre-processing includes recognizing spoken words in the user request and transforming the user request provided as speech user input into a text-based user input. 
     
     
         5 . The method of  claim 3 , wherein the user request is accompanied with one or more attributes, the one or more attributes relating to one or more of the following: current geographical location of the user, currently used software, currently used mobile application, user motions, user movement direction, date, time, and weather information. 
     
     
         6 . The method of  claim 5 , further comprising:
 based on the pre-processing and the one or more attributes, assigning a dialog context label to the user request from a user; and   interpreting the user request using the dialog context label.   
     
     
         7 . The method of  claim 6 , further comprising, based on the interpreting of the user request by the server, applying one of the dialog system rules to the user request. 
     
     
         8 . The method of  claim 6 , further comprising:
 based on determining that the server fails to interpret the user request in isolation, identifying at least one exemplary phrase associated with at least a portion of the user request and the dialog context label; and   based on the identification, applying one of the dialog system rules to the user request.   
     
     
         9 . The method of  claim 1 , wherein the plurality of dialog system rules stipulate instructions on a response to user requests. 
     
     
         10 . A system, comprising:
 a server including at least one processor and a memory storing processor-executable codes, the server being configured to:   maintain a plurality of dialog system rules;   maintain a knowledge database, wherein the knowledge database includes a plurality of intent objects and a plurality of entity objects, wherein the plurality of intent objects and the plurality of entity objects are associated with at least one dialog system rule;   receive an exemplary phrase;   retrieve one or more linguistic elements from the exemplary phrase;   determine that at least one of the linguistic elements is directed to at least one of the plurality of intent objects or the plurality of entity objects; and   add the at least one of the linguistic elements in association with the at least one dialog system rule to the knowledge database.   
     
     
         11 . The system of  claim 10 , wherein the server is further configured to:
 provide the at least one of the linguistic elements for review, wherein the providing includes displaying the plurality of dialog system rules and one or more elements related with the plurality of dialog system rules;   receive one or more corrections to the at least one of the linguistic elements, wherein the one or more corrections include associating the at least one of the linguistic elements with the one or more elements of the plurality of dialog system rules based on user selections; and   update the at least one of the linguistic elements based on the one or more corrections.   
     
     
         12 . The system of  claim 10 , wherein the server is further configured to:
 receive a developer request to create a dialog system rule; and   based on the developer request, create the dialog system rule.   
     
     
         13 . The system of  claim 10 , further comprising:
 a dialog system interface in communication with the server, the system interface being configured to:
 receive a user request from a user; 
 pre-process the user request; and 
 transmit the user request to the server. 
   
     
     
         14 . The system of  claim 13 , wherein the pre-processing includes recognizing spoken words in the user request and transforming the user request provided as speech user input into a text-based user input. 
     
     
         15 . The system of  claim 13 , wherein the user request is accompanied with one or more attributes, the one or more attributes relating to one or more of the following: current geographical location of the user, currently used software, currently used mobile application, user motions, user movement direction, date, time, and weather information. 
     
     
         16 . The system of  claim 15 , wherein the server is further configured to:
 assign a dialog context label to the user request from a user based on the pre-processing and the one or more attributes; and   interpret the user request using the dialog context label.   
     
     
         17 . The system of  claim 16 , wherein one of the dialog system rules is applied to the user request based on the interpreting of the user request by the server. 
     
     
         18 . The system of  claim 16 , wherein the server is further configured to:
 based on determining that the interpretation of the user request in isolation fails, identify at least one exemplary phrase associated with at least a portion of the user request and the dialog context label;   based on the identification, apply one of the dialog system rules to the user request; and   perform one or more actions associated with the one of dialog system rules in response to the user request.   
     
     
         19 . The system of  claim 10 , wherein the plurality of dialog system rules stipulate instructions on a response to user requests. 
     
     
         20 . A non-transitory processor-readable medium having instructions stored thereon, which when executed by one or more processors, cause the one or more processors to implement a method, comprising:
 maintaining a plurality of dialog system rules;   maintaining a knowledge database, wherein the knowledge database includes a plurality of intent objects and a plurality of entity objects, wherein the plurality of intent objects and the plurality of entity objects are associated with at least one dialog system rule;   receiving an exemplary phrase;   retrieving one or more linguistic elements from the exemplary phrase;   determining that at least one of the linguistic elements is directed to at least one of the plurality of intent objects or the plurality of entity objects; and   adding the at least one of the linguistic elements in association with the at least one dialog system rule to the knowledge database.