Back to results
US 12,223,423 B2Granted
Systems and methods for generating automated natural language responses based on identified goals and sub-goals from an utterance
AssigneeCapital One Services, LLC
InventorsAlexandra Coman, Zachary Kulis, Rui Zhang, Liwei Dai, Erik T. Mueller, Vinay Igure
Life of this patent
Estimating the term…
Abstract
The disclosed technology involves autonomously identifying goals and sub-goals from a user utterance and generating responses to the user based on the goals and sub-goals.
Claims20
A system, comprising:
- one or more processors;
- a natural language processing (NLP) device;
- and
- a memory in communication with the one or more processors and storing instructions that, when executed by the one or more processors, are configured to cause the system to:
- continuously receive one or more communications from a user;
- iteratively update one or more programming commands to generate one or more updated programming commands;
- dynamically and autonomously, via the NLP device:
- interpret the one or more communications;
- responsive to interpreting the one or more communications, generate one or more solutions;
- and
- based on the one or more updated programming commands, revise the one or more solutions to generate one or more updated solutions;
- retrieve a list of goals from a domain database;
- identify, using a rule-based approach, a first goal and a first sub-goal in the one or more communications based on the list of goals;
- transform, using the rule-based approach, the one or more communications based on the list of goals to generate one or more normalized communications;
- determine whether the list of goals comprises the first goal and the first sub-goal;
- when the list of goals comprises the first goal and the first sub-goal, determine, using a machine learning model, whether the first sub-goal corresponds to the first goal based on the list of goals;
- and
- when the list of goals comprises the first goal and the first sub-goal and the first sub-goal corresponds to the first goal:
- determine whether the domain database comprises at least one solution plan for the first goal that does not involve the first sub-goal or whether the at least one solution plan can be generated;
- and
- when the domain database comprises the at least one solution plan for the first goal that does not involve the first sub-goal or the at least one solution plan can be generated:
- determine, using the machine learning model, whether the one or more normalized communications comprise an expression of preference for the first sub-goal;
- and
- when the one or more normalized communications comprise the expression of preference for the first sub-goal:
- generate a first recommended solution based on the one or more updated solutions by selecting a solution plan from the at least one solution plan in the domain database or by generating the at least one solution plan;
- determine whether the first recommended solution involves the first sub-goal;
- when the first recommended solution does not involve the first sub-goal, provide the first recommended solution to the user with a first explanation explaining why the first sub-goal was excluded;
- and
- when the first recommended solution involves the first sub-goal, provide the first recommended solution to the user.
26 more elements
A system, comprising:
- one or more processors;
- a natural language processing (NLP) device;
- and
- a memory in communication with the one or more processors and storing instructions that, when executed by the one or more processors, are configured to cause the system to:
- continuously receive one or more first communications from a user;
- iteratively update one or more programming commands to generate one or more updated programming commands;
- dynamically and autonomously, via the NLP device:
- interpret the one or more first communications;
- responsive to interpreting the one or more first communications, generate one or more solutions;
- and
- based on the one or more updated programming commands, revise the one or more solutions to generate one or more updated solutions;
- retrieve a list of goals from a domain database;
- identify, using a rule-based approach, a first goal and a first sub-goal in the one or more first communications based on the list of goals;
- transform, using the rule-based approach, the one or more first communications based on the list of goals to generate one or more normalized communications;
- determine whether the list of goals comprises the first goal and the first sub-goal;
- when the list of goals comprises the first goal and the first sub-goal, determine, using a machine learning model, whether the first sub-goal corresponds to the first goal based on the list of goals;
- and
- when the first goal is in the list of goals, the first sub-goal is in the list of goals, and the first sub-goal corresponds to the first goal:
- determine whether the domain database comprises at least one possible solution for the first goal that does not comprise the first sub-goal or whether the at least one possible solution can be generated;
- and
- when the domain database does not comprise at least one possible solution for the first goal that does not comprise the first sub-goal:
- determine whether the at least one possible solution is admissible;
- and
- when the at least one possible solution is admissible, generate a first recommended solution based on the one or more updated solutions and on the at least one possible solution and provide the first recommended solution to the user.
21 more elements
A system, comprising:
- one or more processors;
- a natural language processing (NLP) device;
- and
- a memory in communication with the one or more processors and storing instructions that, when executed by the one or more processors, are configured to cause the system to:
- receive one or more communications from a user;
- iteratively update one or more programming commands to generate one or more updated programming commands;
- dynamically and autonomously, via the NLP device:
- interpret the one or more communications;
- responsive to interpreting the one or more communications, generate one or more solutions;
- and
- based on the one or more updated programming commands, revise the one or more solutions to generate one or more updated solutions;
- retrieve a list of goals from a domain database;
- identify, using a rule-based approach, a first goal and a first sub-goal in the one or more communications based on the list of goals;
- transform, using the rule-based approach, the one or more communications based on the list of goals to generate one or more normalized communications;
- determine whether the list of goals comprises the first goal and the first sub-goal;
- when the list of goals comprises the first goal and the first sub-goal, determine, using a machine learning model, whether the first sub-goal corresponds to the first goal based on the list of goals;
- and
- when the list of goals comprises the first goal, the list of goals comprises the first sub-goal, and the first sub-goal corresponds to the first goal:
- determine whether the domain database comprises at least one solution plan for the first goal that does not comprise the first sub-goal or whether the at least one solution plan can be generated;
- and
- when the domain database comprises the at least one solution plan for the first goal that does not comprise the first sub-goal or the at least one solution plan can be generated:
- determine, using the machine learning model, whether the one or more normalized communications comprise an expression of preference for the first sub-goal;
- when the one or more normalized communications do not comprise the expression of preference for the first sub-goal:
- generate a recommended solution based on the one or more solutions and by selecting a solution plan from the at least one solution plans or generating a solution plan using an approach such as first-principles planning;
- and
- provide the recommended solution to the user with an explanation of the recommended solution.
23 more elements
Dependencies are inferred from the claim text.