Method, apparatus, and artificial intelligence server for determining artificial intelligence behavior

• US 9,898,700 B2
• Filed: 01/14/2015
• Issued: 02/20/2018
• Est. Priority Date: 11/29/2013
• Status: Active Grant

• Current Assignee: Tencent Technology Company Limited (Tencent Holdings Limited)
• Original Assignee: Tencent Technology Company Limited (Tencent Holdings Limited)
• Inventors: Qiu, Linhui, Wang, Xuxin, Yao, Jianhui, Zhong, Yong, Yang, Min
• Primary Examiner(s): Holmes, Michael B
• Application Number:
  US14/596,663
• Publication Number:
  US 20150154502A1
• Time in Patent Office: 1,133 Days
• Field of Search: 706 47
• CPC Class Codes: G06N 5/048 Fuzzy inferencing

Abstract

According to methods, apparatus and AI servers for determining an AI behavior, a protocol-requesting command sent by an application logic server can be received. The protocol-requesting command can contain an application identifier, a notification message, and current environment data. From a plurality of preset AI systems, an AI system corresponding to the application identifier can be found. The AI system can be formed by a preset plurality of components that include one or more classifying components. From the AI system, a classifying component corresponding to the notification message can be found. The classifying component can be mounted with at least one behavior component. From the classifying component corresponding to the notification message, a behavior component matching the current environment data can be obtained. An AI behavior can be determined based on the obtained behavior component matching the current environment data.

Claims

1. A method for determining an artificial intelligence (AI) behavior, comprising:

receiving a protocol-requesting command sent by an application logic server, wherein the protocol-requesting command contains an application identifier, a notification message, and current environment data;

finding, from a plurality of preset AI systems, an AI system related to the application identifier via looking up and finding a storage structure matching the application identifier from different storage structures, wherein the AI system is formed by a preset plurality of components, the preset plurality of components including one or more classifying components;

finding, from the AI system related to the application identifier, a classifying component related to the notification message, wherein the classifying component is mounted with at least one behavior component;

obtaining, from the classifying component related to the notification message, a behavior component matching the current environment data; and

determining an AI behavior based on the obtained behavior component matching the current environment data.

2. The method according to claim 1, wherein, before finding the AI system related to the application identifier from the plurality of preset AI systems, the method further includes:

encapsulating code for achieving specific functions into components for achieving the specific functions, wherein the components for achieving the specific functions include selecting components, condition components, and behavior components;

selecting a plurality of groups of components from the components for achieving the specific functions, and configuring a classifying component for component managing of each group of the plurality of groups of components; and

forming one AI system including the each group of the plurality of groups of components, and the each group including the classifying component for component managing, the one AI system related to one application program;

obtaining the plurality of AI systems related to a plurality of application programs.

3. The method according to claim 2, wherein:

the classifying component for component managing is mounted with at least one selecting component or at least one condition component;

the at least one selecting component is mounted with at least one behavior component; and

the at least one behavior component is mounted with the at least one condition component.

4. The method according to claim 1, wherein the obtaining of the behavior component matching the current environment data from the classifying component related to the notification message includes:

scanning a plurality of condition components mounted on the classifying component related to the notification message, to determine at least one condition component matching the current environment data;

activating at least one behavior component mounted with the at least one condition component matching the current environment data; and

selecting, by using a selecting component mounted with the at least one behavior component, the behavior component matching the current environment data from the at least one behavior component, based on a filtering rule.

5. The method according to claim 4, wherein before receiving the protocol-requesting command sent by the application logic server, the method further includes:

receiving all application program environment data synchronized with the application logic server; and

wherein the scanning of the plurality of condition components mounted on the classifying component related to the notification message to determine the at least one condition component matching the current environment data includes;

finding, among the all application program environment data, application program environment data related to the at least one condition component of the plurality of condition components, and determining whether the current environment data satisfy the application program environment data related to the at least one condition component of the plurality of condition components; and

when the current environment data satisfy the application program environment data related to the at least one condition component of the plurality of condition components, determining the at least one condition component of the plurality of condition components satisfied with the current environment data as the at least one condition component matching the current environment data.

16. The method according to claim 1, wherein when the notification message includes a message that informing the AI system of playing a card, the application logic server obtains the behavior component from the classifying component of playing basic playing cards to determine the AI behavior of the AI system.

17. The method according to claim 4, wherein the filtering rule comprises a maximum-weight-selection mode, a random-selection mode, a sequential-selection mode, or any other appropriate modes.

6. An apparatus for determining an AI behavior, comprising:

a memory;

a processor coupled to the memory;

a plurality of program modules stored in the memory to be executed by the processor, the plurality of program modules comprising;

a receiving module configured to receive a protocol-requesting command sent by an application logic server, wherein the protocol-requesting command contains an application identifier, a notification message, and current environment data;

a finding module configured to;

find, from a plurality of preset AI systems, an AI system related to the application identifier via looking up and finding a storage structure matching the application identifier from different storage structures, wherein the AI system is formed by a preset plurality of components, the preset plurality of components include one or more classifying components; and

find, from the AI system related to the application identifier, a classifying component related to the notification message, wherein the classifying component is mounted with at least one behavior component;

an obtaining module configured to obtain, from the classifying component related to the notification message, a behavior component matching the current environment data; and

a determining module configured to determine an AI behavior based on the obtained behavior component matching the current environment data.

7. The apparatus according to claim 6, the plurality of program modules further including:

an encapsulating module configured to encapsulate code for achieving specific functions into components for achieving the specific functions, wherein the components for achieving the specific functions include selecting components, condition components, and behavior components;

a configuring module configured to select a plurality of groups of components from the components for achieving the specific functions, and configure a classifying component for component managing of each group of the plurality of groups of components; and

a forming-and-obtaining module configured to;

form one AI system including the each group of the plurality of groups of components and the classifying component for component managing of the each group, the one AI system related to one application program; and

obtain the plurality of AI systems related to a plurality of application programs.

8. The apparatus according to claim 7, wherein:

the classifying component for component managing of the each group of components is mounted with at least one selecting component or at least one condition component;

the at least one selecting component is mounted with at least one behavior component; and

the at least one behavior component is mounted with the at least one condition component.

9. The apparatus according to claim 6, wherein the obtaining module includes:

a scanning unit configured to scan a plurality of condition components mounted on the classifying component related to the notification message;

a determining unit configured to determine at least one condition component matching the current environment data;

an activating unit configured to activate at least one behavior component mounted with the at least one condition component matching the current environment data; and

a selecting unit configured to select, by using a selecting component mounted with the at least one behavior component, the behavior component matching the current environment data from the at least one behavior component, based on a filtering rule.

10. The apparatus according to claim 9, wherein:

the receiving module is further configured to;

receive all application program environment data synchronized with the application logic server; and

the determining unit is configured to;

find, among the all application program environment data, application program environment data related to the at least one condition component of the plurality of condition components, and determine whether the current environment data satisfy the application program environment data related to the at least one condition component of the plurality of condition components; and

when the current environment data satisfy the application program environment data related to the at least one condition component of the plurality of condition components, determine the at least one condition component of the plurality of condition components satisfied with the current environment data as the at least one condition component matching the current environment data.

11. A non-transitory computer-readable medium having computer program for, when being executed by a processor, performing a method for determining an artificial intelligence (AI) behavior, the method comprising:

receiving a protocol-requesting command sent by an application logic server, wherein the protocol-requesting command contains an application identifier, a notification message, and current environment data;

finding, from a plurality of preset AI systems, an AI system related to the application identifier via looking up and finding a storage structure matching the application identifier from different storage structures, wherein the AI system is formed by a preset plurality of components, the preset plurality of components including one or more classifying components;

finding, from the AI system related to the application identifier, a classifying component related to the notification message, wherein the classifying component is mounted with at least one behavior component;

obtaining, from the classifying component related to the notification message, a behavior component matching the current environment data; and

determining an AI behavior based on the obtained behavior component matching the current environment data.

12. The non-transitory computer-readable medium according to claim 11, wherein, before finding the AI system related to the application identifier from the plurality of preset AI systems, the method further includes:

encapsulating code for achieving specific functions into components for achieving the specific functions, wherein the components for achieving the specific functions include selecting components, condition components, and behavior components;

selecting a plurality of groups of components from the components for achieving the specific functions, and configuring a classifying component for component managing of each group of the plurality of groups of components; and

forming one AI system including the each group of the plurality of groups of components and the classifying component for component managing of the each group, the one AI system related to one application program;

obtaining the plurality of AI systems related to a plurality of application programs.

13. The non-transitory computer-readable medium according to claim 12, wherein:

the classifying component for component managing of the each group of components is mounted with at least one selecting component or at least one condition component;

the at least one selecting component is mounted with at least one behavior component; and

the at least one behavior component is mounted with the at least one condition component.

14. The non-transitory computer-readable medium according to claim 11, wherein the obtaining of the behavior component matching the current environment data from the classifying component related to the notification message includes:

scanning a plurality of condition components mounted on the classifying component related to the notification message, to determine at least one condition component matching the current environment data;

activating at least one behavior component mounted with the at least one condition component matching the current environment data; and

selecting, by using a selecting component mounted with the at least one behavior component, the behavior component matching the current environment data from the at least one behavior component, based on a filtering rule.

15. The non-transitory computer-readable medium according to claim 14, wherein before receiving the protocol-requesting command sent by the application logic server, the method further includes:

receiving all application program environment data synchronized with the application logic server; and

wherein the scanning of the plurality of condition components mounted on the classifying component related to the notification message to determine the at least one condition component matching the current environment data includes;

finding, among the all application program environment data, application program environment data related to the at least one condition component of the plurality of condition components, and determining whether the current environment data satisfy the application program environment data related to the at least one condition component of the plurality of condition components; and

when the current environment data satisfy the application program environment data related to the at least one condition component of the plurality of condition components, determining the at least one condition component of the plurality of condition components satisfied with the current environment data as the at least one condition component matching the current environment data.