Information processing system, information processing method and recording medium
By updating the agent's user interface based on user response to the agent's suggestions, the problem of ineffective suggestion presentation in existing technologies is solved, and more efficient suggestion acceptance and behavior change are achieved.
Patent Information
- Application Number
- CN201980048947.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Priority Date
- 2018-07-27
- Filing Date
- 2019-07-22
- Publication Date
- 2024-11-01
- Estimated Expiration
- 2039-07-22
AI Technical Summary
In existing technologies, agent systems fail to effectively consider users' response behavior to suggestions, resulting in ineffective suggestion presentation.
By using information processing systems and methods, the agent's user interface is updated based on whether the user responds to the agent's suggestions. This includes changes to the appearance, sound, and personality of the agent to better present the suggestions.
By adjusting the agent's user interface, user acceptance of suggestions and the effectiveness of behavior modification were improved.
Smart Images

Figure CN112534449B_ABST
Abstract
Description
Technical Field
[0001] The present disclosure relates to an information processing system, an information processing method, and a recording medium. Background Art
[0002] In recent years, agent systems are known that recommend content and actions according to a user's question or request or context using a smartphone, a tablet terminal, a wearable device, or a dedicated terminal such as a home agent.
[0003] For example, Patent Document 1 below discloses a technology that changes the visual aspects of an agent character in a car navigation system to correspond to changes in the user's tendency regarding the agent character's behavior (user preference based on a search for a place or route). In this way, the user can immediately learn of changes in his / her preferences and interests.
[0004] Patent Document 2 listed below discloses a technology for changing the personality parameters of a virtual pet according to the accumulation of transitions between a user and the virtual pet.
[0005] Patent Document 3 listed below discloses a technology that determines the personality of an agent that communicates with a user in a car navigation system based on the user's profile (age, gender, and interests) or surrounding environment (weather and vehicle type).
[0006] Reference List
[0007] Patent Literature
[0008] Patent Document 1: JP 2010-204070A
[0009] Patent Document 2: JP 2002-216026A
[0010] Patent Document 3: JP H11-259446A Summary of the invention
[0011] Technical issues
[0012] However, the technology disclosed in the above patent document does not consider how the agent influences the user. In an agent system that presents suggestions to a user, it is necessary to present the suggestions more effectively by considering whether the user has accepted the suggestions.
[0013] Therefore, the present disclosure proposes an information processing system, an information processing method, and a recording medium, which are capable of presenting suggestions more effectively by updating the user interface of an agent according to the user's behavior modification regarding the agent's suggestions.
[0014] Solution to the problem
[0015] According to the present disclosure, an information processing system is proposed, which includes: a control unit that performs control to update a user interface of an agent according to whether a user has performed a behavior modification in response to a suggestion presented to the user by the agent.
[0016] According to the present disclosure, an information processing method is proposed for causing a processor to perform control to update a user interface of an agent according to whether a user has performed a behavior modification in response to a suggestion presented to the user by the agent.
[0017] According to the present disclosure, a recording medium is proposed on which a program is recorded, the program causing a computer to function as: a control unit that performs control to update an agent's user interface based on whether a user has performed a behavior modification in response to a suggestion presented to the user by the agent.
[0018] Beneficial effects of the present invention
[0019] As described above, according to the present disclosure, suggestions may be presented more effectively by updating the agent's user interface according to user behavior modifications regarding the agent's suggestions.
[0020] Note that the above effects are not necessarily limitative. With or in place of the above effects, any of the effects described in this specification or other effects that can be understood from this specification can be achieved. BRIEF DESCRIPTION OF THE DRAWINGS
[0021] [ Figure 1 ] Figure 1 is a diagram showing an overview of an information processing system according to an embodiment of the present disclosure.
[0022] [ Figure 2 ] Figure 2 is a diagram showing an example of the overall configuration of an information processing system according to the present embodiment.
[0023] [ Figure 3 ] Figure 3 is a block diagram showing an example of the configuration of an information processing apparatus according to the present embodiment.
[0024] [ Figure 4 ] Figure 4 is a diagram showing user parameters used when determining compatibility with a user according to the present embodiment.
[0025] [ Figure 5 ] Figure 5 : is a diagram showing an example of a relationship table between user profile information and user parameters according to the present embodiment.
[0026] [ Figure 6 ] Figure 6 is a diagram showing character parameters used when determining compatibility with a user according to the present embodiment.
[0027] [ Figure 7 ] Figure 7 : is a diagram showing an example of a private character information input screen according to the present embodiment.
[0028] [ Figure 8 ] Figure 8 : is a diagram showing an example of a role parameter assigned to each role according to the present embodiment.
[0029] [ Fig. 9 ] Fig. 9 : is a flowchart showing an example of the flow of the proxy role update process according to the present embodiment.
[0030] [ Fig.10 ] Fig.10 : is a diagram showing an example of parameters for determining attributes of a character according to the present embodiment.
[0031] [ Fig.11 ] Fig.11 FIG. 4 is a diagram showing an example of user parameters assigned based on user profile data according to the present embodiment.
[0032] [ Fig.12 ] Fig.12 : is a flowchart showing an example of an update process flow of step-by-step parameters using an agent character according to the present embodiment.
[0033] [ Fig.13 ] Fig.13 : is a diagram showing a case where the character shape according to the present embodiment is infinitely changed step by step to be updated to the optimal character.
[0034] [ Fig.14 ] Fig.14 : is a flowchart showing an example of an update processing flow using infinite step-by-step changes of agent roles according to the present embodiment.
[0035] [ Fig.15 ] Fig.15 : is a diagram showing an example of the appearance frequency (probability) set for each character according to the present embodiment.
[0036] [ Fig.16 ] Fig.16 : is a flowchart showing an example of the operation flow of the proxy character update process based on the appearance frequency of each character according to the present embodiment. DETAILED DESCRIPTION
[0037] Hereinafter, preferred embodiments of the present disclosure will be described in detail with reference to the accompanying drawings. Note that, in this specification and the drawings, structural elements that have substantially the same function and structure are denoted by the same reference numerals, and repeated explanation of these structural elements is omitted.
[0038] Note that the description will be provided in the following order.
[0039] 1. Overview of Information Processing System According to Embodiments of the Present Disclosure
[0040] 2. Configuration Example of Information Processing Device 2
[0041] 3. Operation processing
[0042] 3-1. Update of agent role
[0043] 3-2. Update based on gradual adjustment of agent roles
[0044] 3-3. Select the agent role for each recommended service
[0045] 3-4. Update based on the frequency of appearance of the agent role
[0046] 4. Conclusion
[0047] 《1. Overview of information processing system according to an embodiment of the present disclosure》
[0048] Figure 1 1 is a diagram showing an overview of an information processing system according to an embodiment of the present disclosure. Figure 1 As shown, the information processing system according to the present embodiment updates the user interface of the agent according to whether the user has performed behavior modification in response to the suggestion of the agent in the agent system, and the agent system provides various pieces of information according to the user's question or request and appropriately presents the suggestion to the user. In this specification, "user interface" refers to the appearance, voice, personality, etc. of the agent that performs interaction with the user.
[0049] For example, if the user interface of the agent presenting the recommendation is such that a nutritionally balanced and healthy diet "Japanese food" is recommended to the user as a dietary suggestion, Figure 1 If a character shown in the upper portion of appears to be behaving poorly, the user may not accept the suggestion (behavior modification does not occur).
[0050] Therefore, in the information processing system according to the present embodiment, the user interface of the agent is updated according to whether the user has accepted the suggestion (specifically, whether the behavior modification is recognized), and the user interface is changed to a (reliable) character in which the user is more likely to accept the suggestion, for example, a character that the user likes, a character that the user admires, or a character that the user sympathizes with. In this way, the suggestion can be presented more effectively.
[0051] For example, as mentioned above, even when Figure 1 When dietary advice is presented to such a character that seems to have poor behavior as shown in the upper part of the figure, and the user does not perform behavioral modification (for example, when even if a nutritionally balanced and healthy diet is recommended to a user who often eats junk food, the eating habits are not modified), the agent character is updated (changed). Specifically, for example, when the person the user admires is a grandmother, the grandmother character is selected from several agent characters, and the user interface (appearance, voice, personality, etc.) of the agent is updated to the grandmother character. When advice is received from the agent of the grandmother character, it can be expected that the user is more likely to accept the advice (for example, behavioral modification of eating habits).
[0052] Here, the agent is a software that autonomously supports the user's behavior, provides (recommends) various pieces of information to the user and interacts with the user, and performs interaction with the user via various devices (agent devices). For example, the agent can be a new agent system (called a master system) that encourages users to change their actions to solve problems from a long-term perspective, and the agent can also be a general agent system that directly provides a session of short-term responses to the user. Here, "master" is the name of a virtual agent that encourages users to spontaneously change their actions (similar to the model of the master). For example, the master system automatically generates predetermined action rules for each specific community or user (means for solving problems in the community, predetermined value standards (i.e., values), predetermined pace of life, etc.), and indirectly encourages the community or user to change actions to solve the community's problems based on the action rules. That is, when the user acts according to the master's words, even if the user is not aware of the action rules, the problems in the community can be solved without being aware of the action rules, the user can take actions that meet the value standards, and the situation of the community is improved.
[0053] Next, we will refer to Figure 2 The overall configuration of the information processing system according to the present embodiment is described. Figure 2 is a diagram showing an example of the overall configuration of an information processing system according to the present embodiment.
[0054] like Figure 2As shown, the information processing system according to the present embodiment includes an agent device 1 that functions as a user interface for presenting suggestions from an agent to a user, and an information processing device 2 that generates an agent character.
[0055] The information processing device 2 obtains advice for the user from an external advice service system (specifically, an advice service server 3), and outputs the advice to the external agent device 1. In this case, the information processing device 2 updates the agent device 1 to the user interface of the selected role agent and presents the advice.
[0056] The advice service server 3 is an example of a service system that generates various pieces of information such as music recommendations or dietary recommendations to be presented to users. Users can enjoy advice from the service system by subscribing to the service system, and expect to provide information from the service system as advice from a predetermined agent role in the agent device 1 used by the user. The information processing device 2 according to the present embodiment can be used as a common basis (platform) that communicates with the advice service server 3 to obtain information (advice) to be presented to the user, and performs control so that the information is presented as advice from an appropriate agent role in the agent device 1 used by the user.
[0057] The service system (advice service server 3) can be an existing information presentation service (for example, a system that recommends content (music, movies, events, games, etc.) and presents advice such as health care) and an agent service, and can be a new agent system called a host system that encourages users to change their actions to solve problems from a long-term perspective.
[0058] The information processing device 2 performs the following control: appropriately selecting (and possibly generating) an agent character; updating the agent device 1 to the user interface of the selected agent character; and causing the agent character to present suggestions. Examples of the updated content of the user interface include changes in the appearance, voice, or personality of the agent character. In addition, the information processing device 2 changes the agent character according to the presence of a behavior modification made by the user in response to suggestions from the agent using the agent device 1, and optimizes the user interface of the agent in the agent device 1. For example, the presence of a behavior modification of the user can be determined based on various pieces of sensing data (camera data, sound data, biometric information, motion sensor information, behavior history, location information, operation history, etc.) acquired from the sensing device 4. The sensing device 4 can be a sensor (surveillance camera, microphone, etc.) set around the user, can be a wearable device worn by the user, and can be set in the agent device 1. A plurality of sensing devices 4 can be set.
[0059] Regarding the change of the appearance of the user interface, for example, when the agent device 1 is a display device such as a smart phone, a tablet terminal, a TV device, an HMD (head mounted display), or a projector device that projects an image, since the agent is displayed in the display screen or the projection area as a virtual agent, the role of the agent can be updated by switching the image. Regarding the change of the appearance, when the agent device 1 is an agent robot made of soft material and having a humanoid appearance, the appearance shape can be changed by adjusting the internal air pressure and protruding a rod-shaped arm from the inside. In addition, when the agent robot has a white and simple shape, the projection of the texture of the agent can be switched by projection from the outside (or inside).
[0060] Regarding the output of the agent's voice, for example, the voiceprint data of the agent character can be sent from the information processing device 2 to the agent device 1, and the agent device 1 speaks by synthesizing the voice from the voiceprint data using a voice synthesis engine. In this case, when updating the character, the information processing device 2 can update the agent's voice by sending the voiceprint data of the character to be updated to the agent device 1.
[0061] The personality of the agent may be reflected in the agent's wording and intonation at the end of a word. For example, when the personality of the agent character is updated, the information processing device 2 sends information about the updated personality of the agent character to the agent device 1, and the agent device 1 presents suggestions using the intonation corresponding to the personality information. The personality information may include information about intonation.
[0062] As described above, the agent device 1 may be a display device such as a smart phone, a tablet terminal, a TV device, or an HMD, may be a projection device that projects an image, and may be a deformable agent robot. In addition, the agent device 1 may be a wearable device such as a smartband, a smart headset, or a smart neck (neck-mounted speaker). In addition, the agent device 1 may be a home voice recognition terminal (home terminal) (so-called voice agent device). The agent device 1 has a function of restoring the agent character selected by the information processing device 2 (display function, sound output function (including a sound synthesis engine), deformation function, etc.), and may present the suggestions received from the information processing device 2 to the user as suggestions from a predetermined agent character.
[0063] In the above, the information processing system according to the embodiment of the present disclosure is described. The configuration of the information processing system according to the present embodiment is not limited to Figure 2 For example, the information processing device 2 may have the function of the suggestion service server 3 , at least some steps of the processing of the information processing device 2 may be performed by the agent device 1 , and the agent device 1 may have all the components of the information processing device 2 .
[0064] Next, a specific configuration of the information processing device 2 included in the information processing system according to the present embodiment will be described with reference to the drawings.
[0065] 《2. Configuration example of information processing device 2》
[0066] Figure 3 2 is a block diagram showing an example of the configuration of the information processing device 2 according to the present embodiment. Figure 3 As shown, the information processing device 2 includes a control unit 200 , a communication unit 210 , and a storage unit 220 .
[0067] <2-1. Control Unit 200>
[0068] The control unit 200 functions as an arithmetic processing device and a control device, and controls the overall operation of the information processing device 2 according to various programs. The control unit 200 is implemented as an electronic circuit such as a CPU (Central Processing Unit) or a microprocessor. The control unit 200 may include a ROM (Read Only Memory) storing programs to be used, calculation parameters, etc., and a RAM (Random Access Memory) temporarily storing parameters that are appropriately changed, etc.
[0069] The control unit 200 according to the present embodiment can function as a user information management unit 201 , a suggested service management unit 202 , an initial role determination unit 203 , a behavior modification determination unit 204 , and a role change unit 205 .
[0070] (User Information Management Unit 201)
[0071] The user information management unit 201 manages profile information input by the user, etc. The profile information of the user is stored in the user profile DB 221. In the present embodiment, the role of the agent is changed based on the behavior modification made by the user in response to the suggestion from the agent to optimize the role (so that the user is more likely to accept the suggestion and effective suggestion can be implemented). On the other hand, for example, as a method of determining the initial role (because no feedback of the behavior modification is obtained from the user), a role compatible with the personality and preference of the user can be selected by using the profile information of the user. Examples of the profile information of the user include gender, date of birth, admired person, and favorite artist type.
[0072] (Suggestion service management unit 202)
[0073] The advice service management unit 202 acquires advice information addressed to the user from the advice service server 3, and distributes the advice information to the agent apparatus 1 used by the user. Information on which user has registered with which advice service server may also be managed.
[0074] (Initial Role Determination Unit 203)
[0075] The initial role determination unit 203 determines an agent role (initial role) initially set to the agent apparatus 1. For example, the initial role determination unit 203 selects a role compatible with the user from agent roles (see the private role DB 222 and the common role DB 223) prepared in advance based on the profile information of the user, and uses the selected role as the initial role. More specifically, for example, various parameters of a role compatible with the user are determined based on the profile information of the user, and these parameters are compared with the role parameters prepared in advance, and a role close to the coordinates is selected as a role compatible with the user.
[0076] (User parameter; UP)
[0077] Here, we will refer to Figure 4 Describes the parameters of the user used when determining compatibility with the user (UP; user parameters). For example, as user parameters, the parameter "Favorite Character Category" of UP0 is set so that "Family and Acquaintances" has a value of 1 (UP0; 1), "Entertainer" has a value of 2 (UP0; 2), "Historical Figures" has a value of 3 (UP0; 3), "Fictional Characters" has a value of 4 (UP0; 4), and "Anime Characters" has a value of 5 (UP0; 5). In addition, the parameter "Favorite Entertainment Type" of UP1 is set so that "Idol" has a value of 1 (UP1; 1), "Singer" has a value of 2 (UP1; 2), "Actor" has a value of 3 (UP1; 3), "TV Actor" has a value of 4 (UP1; 4), and "Comedy" has a value of 5 (UP1; 5).
[0078] According to such parameter setting, a user profile is generated from the profile information registered by the user. Figure 5 The relationship table with user parameters is shown in Figure 1. Figure 5 As shown, parameter values (UP0 to UP4) are applied according to the contents entered in order by the user starting from data 1 (sex) of the profile data. In the table, values are not applied to the blank ("—") portions. When all profile data are entered, all parameters UP0 to UP4 are filled.
[0079] (Character Parameters; CP)
[0080] The initial role determination unit 203 determines the initial role using the parameters (CP; role parameters) of each proxy role and the parameters (UP0 to UP4) based on the profile data of the user. Figure 6Describes the parameters of the user used when determining compatibility with the user (CP; role parameters). For example, as role parameters, the parameter "role category" of CP0 is set so that "family and acquaintances" has a value of 1 (CP0; 1), "entertainer" has a value of 2 (CP0; 2), "historical figure" has a value of 3 (CP0; 3), "fictional character" has a value of 4 (CP0; 4), and "anime character" has a value of 5 (CP0; 5). In addition, the parameter "entertainment type" of CP1 is set so that "idol" has a value of 1 (CP1; 1), "singer" has a value of 2 (CP1; 2), "actor" has a value of 3 (CP1; 3), "TV actor" has a value of 4 (CP1; 4), and "comedy" has a value of 5 (CP1; 5).
[0081] Each item of the parameters of CP and UP has corresponding contents, for example, CP0: character category, and UP0: favorite character category, CP1: entertainment type, and UP1: favorite entertainment type.
[0082] Next, an example of pre-prepared character parameters will be described. In this specification, "character" is used as a general term for the shape (appearance), voice color, or personality used when expressing an agent. In addition, multiple agent characters according to the present embodiment can be prepared in advance on the system side as general characters. General characters include generally known and public characters, such as entertainers who are active on TV and animation, and comic characters. General character information is stored in the general character DB 223. In addition to the general characters, the agent character according to the present embodiment can use people around the user, for example, private characters based on family, friends, acquaintances, etc. (the user's father character or the user's grandmother character). Private character information is stored in the private character DB 222.
[0083] The stored character information includes appearance information (photographic images (2D data and 3D data), illustrations, etc.), sound information (voiceprint data, etc.), personality (personality parameters, etc.), gender, age, and relationship with the user (or close relationship, etc.). It is assumed that the user himself or herself registers information about private characters such as family members and acquaintances that only the user knows. Figure 7 is a diagram showing an example of a private character information input screen. Figure 7 As shown, the user can input and register information about private roles such as family members and acquaintances known only to the user.
[0084] Next, in Figure 8 The role parameters assigned to common roles and private roles are shown in Figure 6 CP information may be stored in each DB as information about each role.
[0085] For example, Figure 8 As shown in Role Example 1 in , values such as "CP0; 1 (Family and acquaintances)", "CP1; - (None)", "CP2; 2 (1970s)" (calculated from date of birth, etc.), "CP3; 5 (Female)", "CP4; 2 (Grandparents)" (see Figure 6 ) is assigned to the private role of "user's grandmother" based on the information about the private role input by the user.
[0086] For example, Figure 8 As shown in the role example 2 in , values such as "CP0; 2 (artist)", "CP1; 2 (singer)", "CP2; 3 (1980s)", "CP3; 1 (male)", "CP4; 5 (other)" (see Figure 6 ) was assigned the generic role of a famous "rock singer".
[0087] (Details of initial role determination)
[0088] As described above, the initial role determination unit 203 determines the initial role using the parameters (CP; role parameters) of each proxy role and the parameters (UP; user parameters) based on the profile data of the user. The initial role determination method is not particularly limited, but for example, the following method is used.
[0089] For example, characters having respective values of CP0 to CP4 are arranged at positions of coordinates ( CP0 , CP1 , CP2 , CP3 , CP4 ) on a multi-dimensional (5-dimensional) space having axes of corresponding parameters.
[0090] The initial character determination unit 203 calculates the distance from the coordinates (UP0, UP1, UP2, UP3, UP4) on the space to the coordinates of each character, and determines the character at the closest coordinates as the initial character.
[0091] For example, the following distance calculation method may be used.
[0092] For n = 0 to 4,
[0093] If UPn=0,
[0094] Then Dn=0,
[0095] If UPn≠0,
[0096] Then Dn=(UPn-CPn)^2,
[0097] Distance D = (D1+D2+D3+D4)^1 / 2
[0098] (Select the character with the smallest distance D as the initial character)
[0099] (Behavior Modification Determination Unit 204)
[0100] The behavior modification determination unit 204 determines the behavior modification performed by the user in response to the suggestion. For example, the behavior modification determination unit 204 may analyze the various pieces of sensing data acquired from the sensing device 4 and determine the user's behavior modification. A specific example of behavior modification determination will be described later with reference to a flowchart.
[0101] (Role Changing Unit 205)
[0102] The role change unit 205 changes (updates) the agent role (user interface) that causes interaction with the user on the agent apparatus 1 according to the behavior modification of the user determined by the behavior modification determination unit 204. For example, if the user does not perform the behavior modification even when the initial role has presented the suggestion a predetermined number of times, the role change unit 205 controls the agent apparatus 1 to select another agent role having similar attributes to the initial role from the private role DB 222 or the common role DB 223, and changes the presented user interface to the user interface of the selected role.
[0103] The change (update) of the agent role is achieved by changing at least one of the appearance, voice, and personality. That is, examples of the change of the agent role include a change of only the appearance, a change of the appearance and voice, a change of only the voice, a change of the voice and personality, a change of the appearance and personality, a change of the personality, and a change of the appearance, voice, and personality. The role change unit 205 can continuously (repeatedly) change the role according to the user's behavior modification to optimize the agent role so that the user performs the behavior modification more (that is, the user accepts the suggestion more effectively). A specific example of the change (update) of the agent role after the initial role is determined will be described later with reference to the flowchart.
[0104] <2-2. Communication Unit 210>
[0105] The communication unit 210 can be connected to an external device via a cable or wirelessly to send data to the external device and receive data from the external device. The communication unit 210 can send data to the suggestion service server 3, the agent device 1, and the sensing device 4 and receive data from the suggestion service server 3, the agent device 1, and the sensing device 4 via a network, for example, via a wired / wireless LAN (Local Area Network), Wi-Fi (registered trademark), Bluetooth (registered trademark), a mobile communication network (LTE (Long Term Evolution)) or 3G (third generation mobile communication method) by communicating with the network.
[0106] <2-3. Storage Unit 220>
[0107] The storage unit 220 is realized as a ROM (Read Only Memory) that stores programs for controlling the processing of the control unit 200 , calculation parameters, and the like, and a RAM (Random Access Memory) that temporarily stores parameters and the like that are changed appropriately.
[0108] For example, the storage unit 220 according to the present embodiment stores a user profile DB (database) 221 , a private role DB 222 , and a common role DB 223 .
[0109] In the above, the configuration of the information processing device 2 according to the present embodiment is described in detail. The information processing device 2 can be implemented as a server on the network, and can be implemented as an edge server having a relatively short communication range and located relatively close to the proxy device 1. The information processing device 2 includes a plurality of devices. In addition, Figure 3 At least some of the components of the information processing device 2 shown may be provided in an external device. In addition, at least some of the functions of the control unit 200 of the information processing device 2 may be implemented by the agent device 1. In addition, the functions of the control unit 200 of the information processing device 2 and the DB of the control unit 200 may be provided in the agent device 1.
[0110] 《3. Operation Processing》
[0111] Next, an operation process of the information processing system according to the present embodiment will be described in detail with reference to the drawings.
[0112] <3-1. Update of Agent Role>
[0113] First, refer to Fig. 9 The updating of the agent role according to the present embodiment is described. In the information processing system according to the present embodiment, the agent to which the initial role is appropriately applied notifies the user of the content of the suggestion generated by each suggestion service. In addition, how the user behaves in response to the notification is detected, and the effect thereof (detection of behavior modification) is measured using various sensing devices 4 such as a camera and a microphone.
[0114] When it is determined based on the measurement result that there is no sufficient effect (ie, there is no behavior modification for the suggestion), the information processing apparatus 2 reselects (updates) the agent role (ie, changes the user interface).
[0115] Will refer to Fig. 9 An example of the flow of the proxy role update process is described in detail. Fig. 9 : is a flowchart showing an example of the flow of the proxy role update process according to the present embodiment.
[0116] like Fig. 9As shown, first, the information processing device 2 selects the initial role determined by the initial role determination unit 203 as the agent (step S103). Specifically, the information processing device 2 sends the role information such as the appearance, voiceprint data and personality of the selected initial role to the agent device 1, so that the agent device 1 uses the initial role to present the suggestion. The suggestion is appropriately obtained by the information processing device 2 from the suggestion service system, and the suggestion is sent to the agent device 1.
[0117] Subsequently, in the information processing device 2, the behavior modification determination unit 204 determines whether the number of times the behavior modification is recognized is less than the expected value t among the N suggestions presented by the selected character (step S106). The number of times the behavior modification is recognized is the number of times the user follows the suggestion. The behavior modification determination unit 204 detects the number of times the user has performed the behavior modification in response to the suggestion based on each piece of sensing data detected by the sensing device 4.
[0118] For example, when the agent device 1 is a home voice recognition terminal (home terminal) and a song is recommended to the user by the terminal, it is determined whether the user has requested the recommended song. For example, when a song is recommended ten times and the number of times the user has requested to play the song is less than a predetermined expected threshold (e.g., 7 times), the behavior modification determination unit 204 reselects (updates) the role.
[0119] For example, when the agent device 1 has presented suggestions on improving eating habits, it is determined whether the user has made a menu according to the suggestions (whether the user has eaten according to the suggestions). For example, when suggestions on menus are presented to the user when selecting meals and the number of times the user selects meals according to the suggestions in ten meals is less than a predetermined expected threshold (e.g., 7 times), the behavior modification determination unit 204 reselects (updates) the role. For example, what kind of meal the user has selected can be input by the user himself or herself in text. The user can use the camera of the agent device 1 to take pictures of food, and the agent device 1 can determine the menu by image analysis. In addition, in the case of an IT-managed company or school cafeteria, the food menu adopted by the user can be determined based on the user's settlement data.
[0120] Subsequently, when the number of recognized behavior modifications is less than the expected value t (step S106: yes), the information processing device 2 sets n=1 (step S109), and the character change unit 205 reselects the character (assigned CP) at the nth coordinates closest to the coordinates of the selected character (in this case, the initial character) (step S112).
[0121] Subsequently, the information processing apparatus 2 determines whether the number of times that behavior modification is recognized among the N times of suggestions presented by the reselected character in the agent apparatus 1 is less than the expected value t (step S115 ).
[0122] When recognizing that the number of behavior modifications is not less than the expected value t (step S115: No), the information processing device 2 determines the reselected character as a new agent (step S121).
[0123] On the other hand, when recognizing that the number of behavior modifications is less than the expected value t (step S115 : Yes), the character change unit 205 increases n by 1 (step S118 ), and selects the character at the nth closest coordinate again (step S112 ).
[0124] By repeating steps S112 to S118 , the role can be set from the initial role in a stepwise manner, and the role can be updated (optimized) to a role with a high suggestion effect for the user (ie, the user is more likely to perform behavior modification (accept suggestion)).
[0125] By executing the processing after step S106 even after a new agent is determined in step S121 , the agent role can be continuously optimized.
[0126] In this embodiment, whether to perform behavior modification can be determined based on the number of times (ratio) behavior modification is recognized when the same suggestion is repeatedly presented by the same agent character. The same suggestion may not necessarily be presented continuously, and the suggestion may be mixed with other suggestions and presented a predetermined number of times within a certain period of time.
[0127] For example, when a user has a bad habit of "biting nails", the information processing device 2 can use the agent character to inform the user of suggestions for changing the habit during the day. When the percentage of effective suggestions (the user stops the behavior according to the suggestions) (the percentage of behavior modification) exceeds 70%, the information processing device 2 can determine that the suggestions presented by the character are effective.
[0128] Duplicate suggestions are not limited to suggestions having exactly the same content, and may be suggestions of the same type or suggestions from a common perspective (ie, suggestions of a common higher category).
[0129] For example, when the user performs behaviors such as "fidgeting" or "smacking tongue" and "biting nails" that are judged as bad habits, the information processing device 2 can repeatedly present suggestions for changing the behaviors using the same agent character. It continuously records whether each "suggestion on bad habits" is effective (the habit is stopped). For example, when the number of times the suggestions are presented reaches 10 times and the number of times the suggestions are effective (the user stops the behavior according to the suggestions) exceeds 7 times, the information processing device 2 determines that the suggestions presented by the character are effective.
[0130] For example, in the case of music recommendation that is not limited to repeated recommendation of the same music, the information processing device 2 repeatedly recommends the same artist or the same type of music to the user a predetermined number of times using the same agent character on the agent device 1. The information processing device 2 determines whether the suggestion presented by the character is valid based on the number of times the user requests to play the recommended music relative to the predetermined number of times (or a predetermined time period) of recommendation (the number of times the recommendation is valid).
[0131] Hereinabove, an example of the operation process according to the present embodiment has been described. Fig. 9 The operation process shown is an example, and the present disclosure is not limited to Fig. 9 Example shown.
[0132] <3-2. Updates based on gradual adjustments to proxy roles>
[0133] Regarding the agent role selection method, although the above-mentioned embodiment uses a method of presetting a plurality of roles, discretely arranging the roles on a multidimensional space formed by parameters (CP; role parameters) that determine role attributes, and modifying the selection of roles based on user profile data or behavior, the present disclosure is not limited thereto.
[0134] For example, a method of adjusting the parameters of the character in a stepwise manner and searching for the best character may be used instead of selecting one character from the unique characters. In this case, the information processing device 2 may sequentially set the proxy character in a stepwise manner to have at least one of the appearance, voice, and personality different from that of the proxy character. Fig.10 As shown, examples of parameters for determining role attributes are defined. Fig.10 : is a diagram showing an example of parameters for determining character attributes according to the present embodiment.
[0135] For example, Fig.10 As shown, for example, a value of 10 steps from "comic style" to "realism" can be set as "P0: character taste". Similarly, a value of 10 steps from "lively" to "gentle" can be set as "P1: character personality".
[0136] In this way, the parameters for determining the attributes of the character are prepared in advance in a step-by-step manner. Fig.10 Although not shown in the figure, the step-by-step parameters of the voice (voiceprint data) can be defined in this way. For example, the step-by-step parameters of the voiceprint data can be set from "high voice (cute voice, young voice, restless voice, etc.)" to "low voice (heavy voice, calm voice, hoarse voice, etc.)" and from "loud voice" to "quiet voice". In addition, "fast tongue to slow tongue" and "dialect to standard language" can be included in the step-by-step parameters of personality (intonation).
[0137] In this embodiment, as an example, the parameters of the character parameters CP0 to CP4 are determined in 10 steps from 0 to 9. Fig.11 As shown, the user parameters (UP) assigned based on the user's profile data can also be set in ten steps.
[0138] Fig.11 2 is a diagram showing an example of user parameters assigned based on the profile data of the user. The initial role determination unit 203 may determine the initial role based on the parameters (UP) assigned to the profile data registered for the user. Fig.10 Among the 10-step values of the user parameter shown, a character in which a value close to the value of UP assigned based on the profile of the target user is used as the setting value of P0 to P3 is set as the initial character.
[0139] The information processing apparatus 2 can update the initial character in the following manner by modifying the use step-by-step parameters with reference to the user's behavior.
[0140] Fig.12 : is a flowchart showing an example of the flow of the updating process of the stepwise parameters using the agent character according to the present embodiment.
[0141] like Fig.12 As shown, first, the information processing device 2 resets n=0 (step S203), and uses the character parameter Pn (for example, Fig.10 any one of P0 to P3 shown) (for example, Fig.10 The suggestions are presented N times (with any one of the values 0 to 9 of the step parameter shown) and the number of times the behavior modification is identified is measured (step S206).
[0142] Subsequently, the information processing apparatus 2 changes the setting value to u-1 for the same parameter Pn, presents the suggestion N times, and measures the number of times the behavior modification is recognized (step S209 ).
[0143] Subsequently, the information processing apparatus 2 changes the setting value to u+1 for the same parameter Pn, presents the suggestion N times, and measures the number of times the behavior modification is recognized (step S212 ).
[0144] Subsequently, the information processing device 2 sets the most effective setting value as a new setting value of the parameter Pn based on the measured number of times the behavior modification is recognized (step S215 ).
[0145] Then, the information processing device 2 adds 1 to n (step S218), and repeats steps S206 to S218 until n=4 (in this example, because there is Fig.10 Parameters P0 to P3 shown) (step S221).
[0146] In this way, by testing the step-by-step values before and after the current setting value of each parameter, the agent can be gradually converted into a character with a higher effect of accepting suggestions. That is, the information processing device 2 can optimize at least one of the appearance, voice and personality of the agent character.
[0147] (Infinite progressive updates of proxy roles)
[0148] For example, in the above example, a method of selecting a role from a plurality of roles prepared in advance and a method of selecting and setting the value of a role parameter from the values of a limited number of steps such as ten steps have been described. However, the present disclosure is not limited thereto.
[0149] The present disclosure is not limited to the method of selecting the character "most obviously recognized to behavior modification" from a limited number of characters, but can continuously change the shape of the character in the form of inserting a limited number of characters. By enabling the user to select a character at an arbitrary position on a continuous line, the character can be changed in an unlimited number of steps and a more optimized character can be selected.
[0150] For example, the creation of continuous changes in shape can be achieved using a path framework method for interpolation and a machine learning method such as GAN (Generative Adversarial Network). In addition, when a machine learning method such as GAN is used, the same algorithm can be applied to the voice (sound color) of the character as well as the shape of the character.
[0151] In this way, for example, Fig.13 As shown, the shape of the character can be continuously changed from the initial character (character A) to the character closest to the initial character (character B), and any shape can be selected.
[0152] In the following, reference will be made to Fig.14 Provide a detailed description. Fig.14 : is a flowchart showing an example of the flow of an update process using infinite step-by-step changes of an agent role according to the present embodiment.
[0153] like Fig.14 As shown, first, the initial role determination unit 203 of the information processing device 2 selects the initial role as the agent (step S253). Regarding the selection of the initial role, for example, as described above with reference to Figures 4 to 8 As described, using the parameter (CP) of each proxy character and the parameter (UP) based on the profile data of the user, the character having the smallest distance D on the multidimensional (5-dimensional) space having axes of the parameters is determined as the initial character.
[0154] Subsequently, the behavior modification determination unit 204 of the information processing device 2 determines whether the number of times a behavior modification is recognized is less than an expected value t among the N times of advice presented by the selected character (step S256 ).
[0155] Subsequently, when the number is less than the expected value t (step S256: yes), the character change unit 205 of the information processing device 2 sets n=1 (step S259), and selects the character at the nth coordinate closest to the coordinates of the selected character (in the multidimensional space of each parameter) (step S262).
[0156] Subsequently, the character changing unit 205 of the information processing device 2 performs image interpolation using the shapes of the two characters to generate an image whose shape changes continuously (so-called image deformation) (step S265), and extracts an image with an arbitrary number of steps on the continuously changing line (step S268).
[0157] Subsequently, the character changing unit 205 of the information processing device 2 sets m=1 (step S271) and applies the image of the mth step as the shape of the character (step S274).
[0158] Subsequently, the role change unit 205 of the information processing device 2 determines whether the number of times the behavior modification is recognized is less than the expected value t among the N times of advice presented by the applied role (step S277 ).
[0159] Subsequently, when the number of times is less than the expected value t (step S277: Yes), the role change unit 205 of the information processing device 2 increases m by 1 (step S280), and repeats steps S274 to S280 until the number of steps reaches the upper limit (step S283).
[0160] When the number of times reaches the expected value t or more (step S277: No), the role change unit 205 of the information processing device 2 selects the applied role as a new agent (step S286).
[0161] In this way, it is possible to generate an image whose shape continuously changes, and select an arbitrary image as an agent character.
[0162] In this example, the change of the character shape has been described. However, the present disclosure is not limited to this. Similarly, for other parameters such as the voice quality (voiceprint data) and personality (intonation) of the character, data in which the parameters are continuously changed can be generated, and any data can be selected as the agent character.
[0163] <3-3. Select agent roles for each recommended service>
[0164] In the above-described embodiment, although the same agent role is selected regardless of the content of the suggestion, the present disclosure is not limited thereto, and a more effective role may be selected for each suggestion service.
[0165] For example, the information processing device 2 can determine the user's behavior modification regarding music recommendation and diet advice, and selectively use the optimal role (having a higher effect of accepting advice) for the music recommendation service and diet advice service based on the determination result of the behavior modification.
[0166] In this way, for example, in the case of music recommendation, since a young and good-looking character of the opposite sex is easily accepted by the user and more effective, the user interface is changed to the corresponding character to present the recommendation. In addition, in the case of diet recommendation, since a familiar grandmother character of the user himself is easily accepted by the user, the user interface is changed to the character to present the recommendation.
[0167] <3-4. Updates based on the frequency of appearance of proxy characters>
[0168] The information processing system according to the present disclosure may use a method of setting the appearance frequency (probability) of each character and changing (updating) the proxy character based on the appearance frequency.
[0169] The information processing device 2 sets the appearance frequency (probability) of each of all the characters (for example, a total of M) prepared in advance. Fig.15 An example of the appearance frequency (probability) set for each character is shown in . All characters appear according to the set appearance frequency.
[0170] In the above embodiment, the following operations are repeatedly performed: one role that is optimal at that point in time is selected, the evaluation is repeatedly performed (e.g., N times) using the selected role for a predetermined period of time, and then the role is reselected (updated). However, in the present embodiment, roles may appear based on the frequency of appearance when presenting suggestions and the evaluation may be performed. For example, when a certain role m appears and a certain suggestion is presented to the user, and the user performs a behavior modification in response to the suggestion, as shown in FIG. Fig.15 As shown, the information processing device 2 performs adjustment (reflection of evaluation) so that the appearance frequency of the character m increases and the appearance frequencies of other characters decrease.
[0171] In this way, it is possible to reflect the evaluation of all roles while showing various roles based on the appearance frequency (probability), and increase the appearance frequency of roles that the user is likely to accept more effectively.
[0172] In the following, reference will be made to Fig.16 Provide a detailed description. Fig.16: is a flowchart showing an example of the flow of operations of a proxy character update process based on the appearance frequency of each character according to the present embodiment.
[0173] like Fig.16 As shown, first, the initial role determination unit 203 of the information processing device 2 selects an initial role as a proxy, and applies a high appearance frequency i[%] (for example, 70%) (step S303).
[0174] Subsequently, the initial role determination unit 203 of the information processing device 2 applies the same frequency of appearance to roles other than the role selected as the initial role (step S306). For example, when the total number of roles is M, the frequency of appearance of (100-i) / (M-1)[%] is applied to roles other than the initial role (for example, if M=31, then (100-70) / (31-1)=1%).
[0175] Subsequently, the role change unit 205 of the information processing device 2 determines the role m to appear according to the appearance frequency of each role, and presents suggestions to the user using the determined role m in the agent device 1 (step S309). This process can be performed when it is necessary for the information processing device 2 to obtain suggestions addressed to the user from the suggestion service and present the suggestions to the user.
[0176] Subsequently, the behavior modification determination unit 204 of the information processing device 2 acquires sensing data from the sensing device 4 and measures the behavior modification of the user (step S312 ).
[0177] Subsequently, the behavior modification determination unit 204 of the information processing apparatus 2 determines whether an intended behavior modification (the user follows the advice) is recognized in response to the advice presented by the agent character (step S315 ).
[0178] Subsequently, when the expected behavior modification is recognized (step S315: Yes), the role change unit 205 of the information processing device 2 increases the appearance frequency of the role m by a specified value a% (step S318). For example, when the previous appearance frequency of the role m is Qm, the new appearance frequency Qm' is Qm'=(1+a / 100)*Qm.
[0179] For example, when the designated value is 10% and Qm is 70%, the new occurrence frequency Qm' is 77%.
[0180] Subsequently, the role change unit 205 of the information processing device 2 reduces the appearance frequency of other roles n (n = 1 to M, n ≠ m) according to the previous appearance frequency of each role (step S321). For example, when the previous appearance frequency of role n is Qn, the new appearance frequency Qn' is Qn' = Qn-(a*Qm / 100)*(Qn / (100-Qm)).
[0181] For example, a 7% increase in the appearance frequency of character m is distributed according to the appearance frequencies of the other 30 characters (when the total number M is 31, the number of characters n other than character m is 30), and the appearance frequencies of other characters are reduced by the distributed increase.
[0182] On the other hand, when the expected behavior modification is not recognized (step S315: No), the role change unit 205 of the information processing device 2 reduces the appearance frequency of the role m by the specified value b% (step S324). For example, when the previous appearance frequency of the role m is Qm, the new appearance frequency Qm' is Qm'=(1-b / 100)*Qm.
[0183] For example, when the designated value is 20% and Qm is 70%, the new occurrence frequency Qm' is 56%.
[0184] The role change unit 205 of the information processing device 2 increases the appearance frequency of other roles n (n=1 to M, n≠m) according to the previous appearance frequency of each role (step S327). For example, when the previous appearance frequency of role n is Qn, the new appearance frequency Qn' is Qn'=Qn+(b*Qm / 100)*(Qn / (100-Qm)).
[0185] For example, a 14% decrease in the appearance frequency of character m is distributed according to the appearance frequencies of the other 30 characters, and the appearance frequencies of the other characters are increased by the distributed decrease.
[0186] 《4. Conclusion》
[0187] As described above, in the information processing system according to the embodiment of the present disclosure, suggestions can be presented more effectively by updating the user interface of the agent according to the behavior modification of the user regarding the agent suggestion.
[0188] Although the preferred embodiments of the present disclosure have been described above with reference to the accompanying drawings, the present disclosure is not limited to the above examples. Those skilled in the art can find various changes and modifications within the scope of the technical concept described in the claims, and it should be understood that these changes and modifications will naturally fall within the technical scope of the present disclosure.
[0189] For example, there is provided a computer program for causing hardware such as a CPU, a ROM, and a RAM included in the information processing apparatus 2 to execute the functions of the information processing apparatus 2. There is also provided a computer-readable storage medium having the computer program stored therein.
[0190] In addition, the effects described in this specification are only illustrative or exemplary effects, and are not restrictive. That is, using the above effects or replacing the above effects, the technology according to the present disclosure can achieve other effects that are clear to those skilled in the art based on the description of this specification.
[0191] Note that the following configurations also fall within the technical scope of the present disclosure. (1)
[0193] An information processing system, comprising:
[0194] A control unit performs control to update a user interface of the agent according to whether the user has performed a behavior modification in response to a suggestion presented to the user by the agent. (2)
[0196] The information processing system according to (1), wherein:
[0197] The control unit performs control to optimize a user interface of the agent so that the user performs behavior modification more in response to the suggestion. (3)
[0199] The information processing system according to (2), wherein:
[0200] When the behavior modification performed by the user in response to the suggestion is lower than a predetermined expected value,
[0201] The control unit performs control to change the agent to another agent whose at least one of appearance, voice, and personality is different from at least one of appearance, voice, and personality of the agent. (4)
[0203] The information processing system according to (2) or (3), wherein:
[0204] The control unit performs control so as to be able to sequentially set, in a step-by-step manner, an agent having at least one of an appearance, a voice, and a personality that is different from at least one of the appearance, the voice, and the personality of the agent, and detects a result of the behavior modification of the user when each agent presents advice to the user, and optimizes at least one of the appearance, the voice, and the personality of the agent based on the detection result. (5)
[0206] The information processing system according to any one of (2) to (4), wherein:
[0207] The control unit performs control to optimize a user interface of the agent for each type of the suggestion. (6)
[0209] The information processing system according to (3) or (4), wherein:
[0210] The control unit optimizes the agent by integrating results of the user's behavior modifications in response to each suggestion when the agent presents suggestions having at least the same higher category a predetermined number of times. (7)
[0212] The information processing system according to any one of (2) to (6), wherein:
[0213] The control unit performs control to:
[0214] generating data that continuously changes at different parts between a plurality of agents, wherein at least one of appearance, voice, and personality of the agents in the plurality of agents is different, extracting data in the middle of the change having an arbitrary number of steps on the generated continuously changing line, and setting the agents in a stepwise manner within the extracted number of steps, and
[0215] Detecting results of the user's behavior modification when each agent presents suggestions to the user, and optimizing at least one of the agent's appearance, voice, and personality based on the detection results. (8)
[0217] The information processing system according to any one of (1) to (7), wherein:
[0218] The control unit sets an initial agent as a user interface of the agent. (9)
[0220] The information processing system according to (8), wherein:
[0221] The control unit sets a user interface of the initial agent according to the profile of the user. (10)
[0223] The information processing system according to (8), wherein:
[0224] The control unit performs control to:
[0225] Setting a relatively high appearance probability for the user interface of the initial agent,
[0226] setting a relatively low appearance probability for user interfaces of agents other than the initial agent,
[0227] configuring a user interface of the agent according to the probability of occurrence, and
[0228] When a behavior modification of the user is identified in response to a suggestion to the user via the set user interface of the agent, an appearance probability of the set user interface of the agent is increased and an appearance probability of user interfaces of other agents is decreased. (11)
[0230] The information processing system according to any one of (1) to (8), wherein:
[0231] The control unit updates the user interface of the external agent device. (12)
[0233] The information processing system according to (11), wherein:
[0234] The control unit performs control to cause the user's agent device to present the advice to the user based on the advice for the user acquired from an external service, and
[0235] The control unit performs control to update a user interface of the agent device according to whether the user has performed a behavior modification in response to the presented suggestion. (13)
[0237] An information processing method, the information processing method comprising:
[0238] The processor performs control to update a user interface of the agent based on whether a user has performed a behavior modification in response to a suggestion presented to the user by the agent. (14)
[0240] A recording medium having recorded thereon a program for causing a computer to function as:
[0241] A control unit performs control to update a user interface of the agent according to whether a user has performed a behavior modification in response to a suggestion presented to the user by the agent.
[0242] Reference numerals list
[0243] 1: Proxy device
[0244] 2: Information processing device
[0245] 3: Recommended service server
[0246] 4: Sensing device
[0247] 200: Control unit
[0248] 201: User information management unit
[0249] 202: Recommended Service Management Unit
[0250] 203: Initial role determination unit
[0251] 204: Behavior Modification Determination Unit
[0252] 205: Role Change Unit
[0253] 210: Communication unit
[0254] 220: Storage unit
[0255] 221: User profile DB
[0256] 222: Private Role DB
[0257] 223: General Role DB
Claims
1. An information processing system, comprising: a control unit configured to perform control to update a user interface of the agent according to whether the user has performed a behavior modification in response to a suggestion presented to the user by the agent, Wherein, the control unit is further configured to perform control when the behavior modification performed by the user in response to the suggestion is lower than a predetermined expected value, so as to: changing the agent to another agent, at least one of the appearance, voice, and personality of the other agent being different from at least one of the appearance, voice, and personality of the agent; detecting results of the user's behavior modification when each agent presents a suggestion to the user; and At least one of the appearance, voice, and personality of the agent is optimized based on the detection results.
2. The information processing system according to claim 1, wherein: The control unit is further configured to perform control to enable sequentially setting an agent having at least one of appearance, voice, and personality different from at least one of appearance, voice, and personality of the agent in a stepwise manner.
3. The information processing system according to claim 1, wherein: The control unit is further configured to perform control to optimize a user interface of the agent for each type of the suggestion.
4. The information processing system according to claim 1, wherein: The control unit is further configured to optimize the agent by integrating results of the user's behavior modification in response to each suggestion when the agent presents suggestions having at least the same higher category a predetermined number of times.
5. The information processing system according to claim 1, wherein: The control unit is further configured to perform control to: Generate data that continuously changes in different parts between multiple agents, where at least one of the appearance, voice and personality of the agents in the multiple agents is different, extract data in the middle of the change with an arbitrary number of steps on the generated continuously changing line, and set the agents in a step-by-step manner within the extracted number of steps.
6. The information processing system according to claim 1, wherein: The control unit is further configured to set an initial agent as a user interface of the agent.
7. The information processing system according to claim 6, wherein: The control unit is further configured to set a user interface of the initial agent according to the profile of the user.
8. The information processing system according to claim 6, wherein: The control unit is further configured to perform control to: Setting a relatively high appearance probability for the user interface of the initial agent, setting a relatively low appearance probability for user interfaces of agents other than the initial agent, configuring a user interface of the agent according to the probability of occurrence, and When a behavior modification of the user is identified in response to a suggestion to the user via the set user interface of the agent, the appearance probability of the set user interface of the agent is increased and the appearance probability of user interfaces of other agents is decreased.
9. The information processing system according to claim 1, wherein: The control unit is further configured to perform control to update a user interface of the external agent device.
10. The information processing system according to claim 9, wherein: The control unit is further configured to perform control to cause the user's agent device to present the suggestion to the user based on the suggestion for the user acquired from the external service, and The control unit performs control to update a user interface of the agent device according to whether the user has performed a behavior modification in response to the presented suggestion.
11. An information processing method, the information processing method comprising: controlling, by a processor, to update a user interface of the agent based on whether a user has performed a behavior modification in response to a suggestion presented to the user by the agent, The method further comprises, when the behavior modification performed by the user in response to the suggestion is lower than a predetermined expected value: changing, by a processor, the agent to another agent, at least one of the appearance, voice, and personality of the other agent being different from at least one of the appearance, voice, and personality of the agent; detecting, by a processor, a result of the user's behavior modification when each agent presents a suggestion to the user; and At least one of the appearance, voice, and personality of the agent is optimized by a processor based on the detection results.
12. A recording medium having recorded thereon a program for causing a computer to function as: a control unit configured to perform control to update a user interface of the agent according to whether the user has performed a behavior modification in response to a suggestion presented to the user by the agent, in, The control unit is further configured to perform control when the behavior modification performed by the user in response to the suggestion is below a predetermined expected value, to: changing the agent to another agent, at least one of the appearance, voice, and personality of the other agent being different from at least one of the appearance, voice, and personality of the agent; detecting results of the user's behavior modification when each agent presents a suggestion to the user; and At least one of the appearance, voice, and personality of the agent is optimized based on the detection results.
