Expandable service architecture with configurable dialogue manager

Patent metadata

Field	Value
Publication number	US-10977319-B2
Application number	US-201715645641-A
Country	US
Kind code	B2
Filing date	Jul 10, 2017
Priority date	Jul 10, 2017
Publication date	Apr 13, 2021
Grant date	Apr 13, 2021

Abstract

Methods, systems, and computer programs are presented for facilitating user engagement with a bot. One example method includes receiving, by a dialogue manager server, a sequence specification for a user activity that identifies a type of interaction between a user and a network service, the sequence specification including instructions for implementing a trigger action in the event of a negative user input into the bot. The negative user input in the sequence of interactions is detected and the trigger action is implemented in response to detecting the negative user input.

First claim

The invention claimed is:

1. A method comprising:
receiving, by a dialogue manager server, a plurality of sequence specifications for a plurality of user activities, the plurality of sequence specifications including a first sequence specification for a first user activity that includes a type of interaction between a client device and a network service, the network service including the dialogue manager server and a plurality of service servers, the first sequence specification comprising a sequence of interactions between the dialogue manager server and the plurality of service servers providing services to implement the first user activity;
processing user input from the client device to detect an intent associated with the user input;
determining that the intent corresponds to the first user activity; and
in response to determining the intent corresponds to the first user activity, executing, by the dialogue manager server, the first sequence specification by invoking the plurality of service servers to provide the services to implement the first user activity in accordance with the sequence of interactions specified by the first sequence specification.

2. The method of claim 1 , wherein the sequence specification includes instructions for implementing a trigger action in the event of a negative user input in the sequence of interactions, and wherein the method further comprises:
detecting the negative user input in the sequence of interactions after executing the sequence specification; and
implementing the trigger action in response to detecting the negative user input.

3. The method of claim 2 , further comprising detecting an expiry of a time (T) and associating the expiry of the time (T) with the negative user input.

4. The method of claim 3 , wherein the negative user input includes a user failure, after the expiry of the time (T), to perform one or more of: answering a question, clicking on a listing in a search result, entering text into a search field or dialogue box, providing a verbal command or comment, and providing a digital image.

5. The method of claim 3 , wherein the time (T) is determined based on output of a machine-learning model using aggregate historical user response times, and one or more of real-time user response times, device type usage, demographic information, and gender data, as training data.

6. The method of claim 3 , wherein the trigger action comprises asking a proactive question or providing a comment after the expiry of the time (T).

7. The method of claim 2 , wherein the trigger action, implemented in response to the detected negative user input, is based on a user intent detected in a prior interaction in the sequence of interactions.

8. A system including a dialogue manager server, the system comprising:
a memory comprising instructions; and
one or more computer processors, wherein the instructions, when executed by the one or more computer processors, cause the one or more computer processors to perform operations comprising:
receiving, by a dialogue manager server, a plurality of sequence specifications for a plurality of user activities, the plurality of sequence specifications including a first sequence specification for a first user activity that includes a type of interaction between a client device and a network service, the network service including the dialogue manager server and a plurality of service servers, the first sequence specification comprising a sequence of interactions between the dialogue manager server and the plurality of service servers providing services to implement the first user activity;
processing user input from the client device to detect an intent associated with the user input;
in response to determining the intent corresponds to the first user activity, executing, by the dialogue manager server, the first sequence specification by invoking the plurality of service servers to provide the services to implement the first user activity in accordance with the sequence of interactions specified by the first sequence specification.

9. The system of claim 8 , wherein the sequence specification includes instructions for implementing a trigger action in the event of a negative user input in the sequence of interactions, and wherein the operations further comprise:
detecting the negative user input in the sequence of interactions after executing the sequence specification; and
implementing the trigger action in response to detecting the negative user input.

10. The system of claim 9 , wherein the operations further comprise detecting an expiry of a time (T) and associating the expiry of the time (T) with the negative user input.

11. The system of claim 10 , wherein the negative user input includes a user failure, after the expiry of the time (T), to perform one or more of: answering a question, clicking on a listing in a search result, entering text into a search field or dialogue box, providing a verbal command or comment, and providing a digital image.

12. The system of claim 10 , wherein the time (T) is determined based on output of a machine-learning model using aggregate historical user response times, and one or more of real-time user response times, device type usage, demographic information, and gender data, as training data.

13. The system of claim 10 , wherein the trigger action comprises asking a proactive question or providing a comment after the expiry of the time (T).

14. The system of claim 9 , wherein the trigger action, implemented in response to the detected negative user input, is based on a user intent detected in a prior interaction in the sequence of interactions.

15. A non-transitory machine-readable storage medium including instructions that, when executed by one or more processors of a machine, cause the machine to perform operations comprising:
receiving, by a dialogue manager server, a plurality of sequence specifications for a plurality of user activities, the plurality of sequence specifications including a first sequence specification for a first user activity that includes a type of interaction between a client device and a network service, the network service including the dialogue manager server and a plurality of service servers, the first sequence specification comprising a sequence of interactions between the dialogue manager server and the plurality of service servers providing services to implement the first user activity;
processing user input from the client device to detect an intent associated with the user input;
determining that the intent corresponds to the user activity; and
in response to determining the intent corresponds to the first user activity, executing, by the dialogue manager server, the first sequence specification by invoking the plurality of service servers to provide the services to implement the first user activity in accordance with the sequence of interactions specified by the first sequence specification.

16. The medium of claim 15 , wherein the sequence specification includes instructions for implementing a trigger action in the event of a negative user input in the sequence of interactions, and wherein the operations further comprise:
detecting the negative user input in the sequence of interactions after executing the sequence specification; and
implementing the trigger action in response to detecting the negative user input.

17. The medium of claim 16 , wherein the operations further comprise detecting an expiry of a time (T) and associating the expiry of the time (T) with the negative user input.

18. The mediu

Assignees

• Ebay Inc

Inventors

• Bright Ido Avigdor

Classifications

• G06Q30/0643

  graphically representing goods, e.g. 3D product representation · CPC title

• G06F16/951 Primary

  Indexing; Web crawling techniques · CPC title

• G06Q10/00

  Administration; Management · CPC title

• G06Q30/0617 Primary

  Representative agent · CPC title

• G06F3/0482

  Interaction with lists of selectable items, e.g. menus · CPC title