Virtual assistant conversations

• US 10,438,610 B2
• Filed: 08/25/2014
• Issued: 10/08/2019
• Est. Priority Date: 01/15/2008
• Status: Active Grant

• Current Assignee: Verint Americas Incorporated (Verint Systems Incorporated)
• Original Assignee: Verint Americas Incorporated (Verint Systems Incorporated)
• Inventors: Brown, Fred, Miller, Tanya M, Zartler, Mark, Brown, Molly Q
• Primary Examiner(s): Utama, Robert J
• Application Number:
  US14/467,715
• Publication Number:
  US 20140365223A1
• Time in Patent Office: 1,870 Days
• CPC Class Codes: G06F 16/3329 Natural language query formulation or dialogue systems
• CPC Class Codes: G06F 16/63 Querying
• CPC Class Codes: G06F 40/40 Processing or translation of natural language natural language analysis G06F40/20; semantic analysis G06F40/30
• CPC Class Codes: G06N 5/02 Knowledge representation; Symbolic representation
• CPC Class Codes: G10L 15/18 using natural language modelling
• CPC Class Codes: G10L 25/48 specially adapted for particular use

Abstract

A virtual assistant may communicate with a user in a natural language that simulates a human. The virtual assistant may be associated with a human-configured knowledge base that simulates human responses. In some instances, a parent response may be provided by the virtual assistant and, thereafter, a child response that is associated with the parent response may be provided.

Claims

1. A method comprising:

under control of one or more processors executing computerized instructions stored in memory;

causing a virtual assistant to be presented via a computing device to enable a conversation with a user in a natural language, wherein the computerized instructions are configured via a graphical user interface that receives features of a knowledge base for storage in the memory, the features of the knowledge base organized to trigger outputs according to units of vocabulary patterns arranged in the features of the knowledge base, wherein the units of vocabulary patterns are stored in the memory with respective labels for each feature;

receiving a speech input string via the virtual assistant during the conversation;

converting the speech input string to converted data by substituting portions of the speech input string with sets of terms bearing corresponding labels that refer back to the respective labels of the knowledge base;

matching the respective labels to the corresponding labels to identify a digital response to the speech input string; and

causing the digital response to be presented to the user in real-time via the virtual assistant.

2. The method of claim 1, further comprising:

receiving the speech input string; and

causing the virtual assistant to provide a natural language response to the speech input string; and

wherein the natural language response for the speech input string is linked in the knowledge base to the converted data via the label.

3. The method of claim 1, further comprising configuring the virtual assistant for human simulation by linking in the knowledge base a series of the units of vocabulary patterns to map the digital response to one or more of the sets of terms in the converted data.

4. The method of claim 1, further comprising obtaining a parameter value of a variable stored in the memory for use in converting the speech input string during the conversation between the user and the virtual assistant,

wherein the parameter value includes at least one of:

a user name;

an email address;

an IP address;

a credit card number;

a user'"'"'s gender;

or

information related to user activity on a site.

9. The method of claim 1, further comprising:

determining a pattern of terms for the converted data, each of the terms of the patterns of terms comprising at least one of;

one or more vocab terms comprising a grouping of at least one of one or more unambiguous synonyms of a word in the speech input or one or more spelling variations of the word in the speech input;

or

one or more wildcards that each function as a placeholder for a word or words.

10. The method of claim 1, further comprising:

wherein the response indicates that the task was performed by the application.

5. One or more non-transitory computer-readable storage media storing instructions that, when executed by one or more processors, cause the one or more processors to perform operations comprising:

causing a virtual assistant to be presented via a computing device to enable a conversation with a user in a natural language, wherein the virtual assistant is configured with instructions stored in computerized memory to access features of a knowledge base stored in a computer readable media, the features of the knowledge base organized to trigger parent responses and child responses according to units of vocabulary patterns arranged in the features of the knowledge base, wherein the vocabulary patterns are stored in the memory with respective labels and the instructions implement the following steps;

receiving a speech input string via the virtual assistant during the conversation;

substituting portions of the speech input string with sets of terms bearing corresponding labels that refer back to the respective labels of the knowledge base;

converting the speech input string to converted data comprising the sets of terms;

matching the respective labels and the corresponding labels to determine a parent response for responding to the speech input string that is received during the conversation;

causing the virtual assistant to provide the parent response; and

based on at least one of a value that is external to the conversation, a value that is obtained from an application, or a value that is obtained from a previous conversation between the user and the virtual assistant, determining a child response for responding to additional user input that is received during the conversation, the child response being accessible via the parent response.

6. The one or more non-transitory computer-readable storage media of claim 5, wherein the parent response is associated with a unit that maps a pattern of concepts that is determined for the user input to the parent response.

7. The one or more non-transitory computer-readable storage media of claim 6, wherein the pattern of concepts includes at least one of an order of the concepts or a proximity of the concepts to each other.

8. The one or more non-transitory computer-readable storage media of claim 5, wherein the parent response is determined based at least in part on one or more values related to at least one of activity of the user on a site or previous user input from the user during the conversation.