Information processing system and information processing method to control behavior of an agent
Provided is an information processing system that includes a user information storage unit that stores user information related to a user, a personality information storage unit that stores personality information defining a personality unique to an agent, and a control unit that controls an externalized behavior which is a behavior externalized by the agent, based on the personality information.
1. An information processing system, comprising:
a memory configured to:
store user information related to a user; and
store personality information that defines a personality unique to an agent, wherein
the personality information includes individuality information that defines individuality of the agent,
the personality information is based on one of a character in a real space or an animated character, and
the personality information is learned by a machine learning algorithm using learning data associated with the personality of the agent; and
a processor configured to:
determine the personality information is to be updated based on a growth stage of the personality of the agent;
acquire new learning data based on the determination that the personality information is to be updated, wherein
the new learning data includes a first set of phrases related to an adolescent stage of the agent in a case where the personality information is associated with a childhood stage of the agent, and
the new learning data includes a second set of phrases related to an adulthood stage of the agent in a case where the personality information is associated with the adolescent stage of the agent;
control the machine learning algorithm to relearn the personality information using the new learning data to update the personality information;
generate a behavior candidate based on the updated personality information and the user information, wherein
the behavior candidate is a candidate of an externalized behavior of the agent, and
the behavior candidate includes an utterance candidate of a plurality of utterance candidates;
calculate a matching degree between a content of the utterance candidate and the individuality information;
select the utterance candidate from the plurality of utterance candidates that has the calculated matching degree higher than a threshold;
transmit information indicating the selected utterance candidate to a user terminal; and
control the user terminal to output the selected utterance candidate as the externalized behavior of the agent.
2. The information processing system according to claim 1 , wherein the personality information is not influenced by a change in the user information.
3. The information processing system according to claim 1 , wherein the personality information is information defined based on an original model related to the agent.
4. The information processing system according to claim 1 , wherein the personality information further includes basic information that defines the externalized behavior performed by the agent.
5. The information processing system according to claim 1 , wherein the processor is further configured to update information related to trends with regard to the personality information.
6. The information processing system according to claim 1 , wherein
the personality information further includes behavior information related to the agent, and
the processor is further configured to update the behavior information.
7. The information processing system according to claim 6 , wherein
the personality information further includes word information related to words uttered by the agent, and
the processor is further configured to update the word information.
8. The information processing system according to claim 1 , wherein, based on a change in status information related to a status of the agent, the processor is further configured to update, triggered by the change, the personality information to first personality information corresponding to the status information after the change.
9. The information processing system according to claim 8 , wherein
the status information includes growth information related to the growth stage of the agent, and
based on a change in the growth information, the processor is further configured to update, triggered by the change, the personality information corresponding to the growth information after the change.
10. The information processing system according to claim 1 , wherein the processor is further configured to update the personality information by addition of additional information that defines a new personality of the agent to the personality information.
11. The information processing system according to claim 1 , wherein the processor is further configured to update the personality information based on event information related to an event in which the user participates.
12. The information processing system according to claim 1 , wherein
the personality information further includes location information that indicates a location related to the agent, and
the processor is further configured to control the externalized behavior based on the location information.
13. The information processing system according to claim 12 , wherein the processor is further configured to control the externalized behavior such that the agent recommends a behavior related to a location of the user.
14. The information processing system according to claim 1 , wherein
the memory is further configured to store a plurality of personality information for a plurality of agents,
the processor is further configured to:
select at least one agent from the plurality of agents based on the plurality of personality information; and
control the externalized behavior of the selected at least one agent based on the personality information related to the selected at least one agent, the plurality of agents includes the agent, and
the plurality of personality information includes the personality information.
15. The information processing system according to claim 14 , wherein the selection of the at least one agent is based on a relationship between the plurality of personality information of the plurality of the agents and information related to a service provided to the user.
16. The information processing system according to claim 1 , wherein the processor is further configured to control a user device to output the externalized behavior in a form of a script.
17. An information processing method, comprising:
storing, by a memory, user information related to a user;
storing, by the memory, personality information that defines a personality unique to an agent, wherein
the personality information includes individuality information that defines individuality of the agent,
the personality information is based on one of a character in a real space or an animated character, and
the personality information is learned by a machine learning algorithm using learning data associated with the personality of the agent;
determining, by a processor, the personality information is to be updated based on a growth stage of the personality of the agent;
acquiring, by the processor, new learning data based on the determination that the personality information is to be updated, wherein
the new learning data includes a first set of phrases related to an adolescent stage of the agent in a case where the personality information is associated with a childhood stage of the agent, and
the new learning data includes a second set of phrases related to an adulthood stage of the agent in a case where the personality information is associated with the adolescent stage of the agent;
controlling, by the processor, the machine learning algorithm to relearn the personality information using the new learning data to update the personality information;
generating, by the processor, a behavior candidate based on the updated personality information and the user information, wherein
the behavior candidate is a candidate of an externalized behavior of the agent, and
the behavior candidate includes an utterance candidate of a plurality of utterance candidates;
calculating, by the processor, a matching degree between a content of the utterance candidate and the individuality information;
selecting, by the processor, the utterance candidate from the plurality of utterance candidates that has the calculated matching degree higher than a threshold;
transmitting information indicating the selected utterance candidate to a user terminal; and
controlling, by the processor, the user terminal to output the selected utterance candidate as the externalized behavior of the agent.
18. A non-transitory computer-readable medium having stored thereon, computer-executable instructions which, when executed by an information processing system, cause the information processing system to execute operations, the operations comprising:
storing, by a memory, user information related to a user;
storing, by the memory, personality information that defines a personality unique to an agent, wherein
the personality information includes individuality information that defines individuality of the agent,
the personality information is based on one of a character in a real space or an animated character, and
the personality information is learned by a machine learning algorithm using learning data associated with the personality of the agent;
determining, by a processor, the personality information is to be updated based on a growth stage of the personality of the agent;
acquiring, by the processor, new learning data based on the determination that the personality information is to be updated, wherein
the new learning data includes a first set of phrases related to an adolescent stage of the agent in a case where the personality information is associated with a childhood stage of the agent, and
the new learning data includes a second set of phrases related to an adulthood stage of the agent in a case where the personality information is associated with the adolescent stage of the agent;
controlling, by the processor, the machine learning algorithm to relearn the personality information using the new learning data to update the personality information;
generating, by the processor, a behavior candidate based on the updated personality information and the user information, wherein
the behavior candidate is a candidate of an externalized behavior of the agent, and
the behavior candidate includes an utterance candidate of a plurality of utterance candidates;
calculating, by the processor, a matching degree between a content of the utterance candidate and the individuality information;
selecting, by the processor, the utterance candidate from the plurality of utterance candidates that has the calculated matching degree higher than a threshold;
transmitting information indicating the selected utterance candidate to a user terminal; and
controlling, by the processor, the user terminal to output the selected utterance candidate as the externalized behavior of the agent.