Preferential treatment determination system

Abstract

To make the selection of passengers eligible for preferential treatment more appropriate. [Solution] The preferential treatment determination system S, which analyzes multiple passengers U, comprises: a proposal unit 112 installed in a vehicle V and providing services to passengers U; an estimation unit 111 that estimates a confidence level C1 between passenger U and the vehicle V, the service, or the manufacturing brand of the vehicle V based on the passenger U's response to the offer from the proposal unit 112; a first calculation unit 524 that calculates a value of a first index K1, which is defined to increase or decrease according to the confidence level C1, based on the confidence level C1 estimated by the estimation unit 111; a preferential treatment determination unit 541 that scores the preferential treatment level N1 of the passenger U who was the target of the estimation of the first index K1, based on the value of the first index K1 calculated by the first calculation unit 524; and a candidate selection unit 542 that selects and outputs preferred customers Us from multiple passengers U based on the level of the preferential treatment level N1 scored by the preferential treatment determination unit 541.

Description

Technical Field

[0001] The present disclosure relates to a preferential determination system for determining passengers of a moving object.

Background Art

[0002] Patent Document 1 discloses, as an example of a preferential determination system, a sales promotion prediction device that selects candidates having an influence on transmission for a target product. This sales promotion prediction device includes an operation information acquisition unit, a sales promotion prediction unit, and an output unit.

[0003] Here, the operation information acquisition unit acquires operation information for post data related to the target product among the post data posted by candidates. The sales promotion prediction unit predicts the degree of sales promotion of the target product by the candidates' posts based on the acquired operation information. The output unit outputs the result of the degree of sales promotion predicted by the sales promotion prediction unit.

[0004] Patent Document 2 discloses, as another example of a preferential determination system, an influence degree analysis device that analyzes a useful influence degree for evaluating whether a user is a useful influencer for a promotion user who conducts a promotion. This influence degree analysis device calculates a useful influence degree based on an influence degree and a fan degree.

[0005] Here, the influence degree indicates the degree of influence that each user participating in the communication service gives to other users. The fan degree indicates the degree of favor of each user for the promotion user. The useful influence degree is used to evaluate whether a user is a useful influencer for a promotion user.

Prior Art Documents

Patent Documents

[0006]

Patent Document 1

Patent Document 2

[0007] The inventors of this application sought to select passengers who would be eligible for various preferential treatments, such as test rides of the mobile device, in order to promote the product. To meet this need, it is conceivable to adopt conventional marketing methods such as those disclosed in Patent Document 1 or 2. By adopting conventional marketing methods, it is possible to select influencers (candidates) with the ability to disseminate information. It has been thought that selecting candidates with the ability to disseminate information would contribute to the promotion of the product.

[0008] However, traditional marketing methods do not consider the strength of the connection between the product (such as a mobile device), the services provided by the product, or the brand that manufactures the product when selecting candidates.

[0009] Therefore, if a candidate is selected who is considered to have a relatively weak connection to the target product, such as a heavy user of a competing product, even if preferential treatment is given to that candidate, it may only amount to superficial PR and may not necessarily lead to effective PR.

[0010] This disclosure is made in view of the above, and its purpose is to enable a more appropriate selection of passengers who are eligible for preferential treatment. [Means for solving the problem]

[0011] A first aspect of this disclosure relates to a preferential treatment determination system for analyzing multiple passengers, each riding in a mobile vehicle. The preferential treatment determination system includes: a service provision unit mounted on the mobile vehicle and providing services to the passengers; an estimation unit that estimates the degree of trust between the passenger and the mobile vehicle, the service, or the manufacturing brand of the mobile vehicle based on the passenger's response to the services provided by the service provision unit; a first calculation unit that calculates a value of a first indicator defined to increase or decrease according to the degree of trust estimated by the estimation unit; a preferential treatment determination unit that scores the degree of preferential treatment of the passengers who are the target of estimation for the first indicator among the multiple passengers based on the value of the first indicator calculated by the first calculation unit; and a candidate selection unit that selects and outputs passengers who are to be given preferential treatment from among the multiple passengers based on the level of the preferential treatment value scored by the preferential treatment determination unit.

[0012] In recent years, there has been an increase in mobile vehicles equipped with functions that provide various services, such as suggesting destinations tailored to the passenger's purpose, suggesting travel routes to those destinations, and suggesting music to be played inside the vehicle.

[0013] On the other hand, among ordinary passengers, including vehicle drivers, there may be some who feel aversion to services provided by mobile vehicles. Such aversion is reflected in the passengers' responses to services provided by mobile vehicles, such as the frequency with which they accept or reject the services offered. The inventors of this application believe that this aversion fluctuates depending on the level of trust between the passenger and the mobile vehicle, the service, or the brand of manufacture of the mobile vehicle. If a high level of trust has been established between the passenger and the manufacturer's brand, it is thought that the aversion to the services provided will also be mitigated in proportion to that trust.

[0014] In other words, the inventors of this invention believe that each passenger has their own level of trust in the vehicle, service, or brand of manufacture, and that this level of trust is reflected in the passenger's response to the services provided by the vehicle.

[0015] Furthermore, the higher the level of trust, the stronger the "connection" between the passenger and the vehicle, the service provided, or the brand being manufactured. The strength of this connection can be quantified according to the level of trust. By giving preferential treatment to passengers with strong connections, it becomes possible to achieve more effective PR compared to passengers with weaker connections.

[0016] Specifically, according to the first embodiment, the preferential treatment determination system estimates the level of confidence based on the passenger's response to the services provided, and calculates a value for a first indicator indicating the strength of the connection based on that level of confidence. The preferential treatment determination system then scores the passenger's level of preferential treatment based on the value of the first indicator, and selects passengers based on that level of preferential treatment. This configuration allows for a more appropriate selection of passengers who are eligible for preferential treatment.

[0017] Furthermore, according to a second aspect of this disclosure, the preferential treatment determination system may include a boarding history acquisition unit that acquires the boarding history of the passenger based on electrical signals input from the onboard equipment of the mobile body, and the first calculation unit calculates the value of the first index by referring to the reliability and the boarding history acquired by the boarding history acquisition unit, so as to increase or decrease according to the reliability and the boarding history.

[0018] According to the second embodiment described above, by referring to the passenger's boarding history in addition to the reliability score, the "strength of the passenger's attachment" to the mobile device can be reflected when calculating the first indicator. This is advantageous in more appropriately selecting passengers who are eligible for preferential treatment.

[0019] Furthermore, according to a third aspect of this disclosure, the first calculation unit may calculate the value of the first indicator by referring to at least one of the following as the boarding history: the boarding frequency of the boarding, the cumulative boarding time of the boarding, and the cumulative distance traveled by the boarding vehicle.

[0020] According to the third aspect, by referring to the boarding frequency, boarding time, and travel distance, when calculating the value of the first index, the "strength of attachment" to the moving object can be more appropriately reflected. This is advantageous for more appropriately selecting the passengers to be given preferential treatment.

[0021] Also, according to the fourth aspect of the present disclosure, the preferential treatment determination system includes a communication unit connected to the Internet and a relationship acquisition unit that acquires relationship information characterizing the depth of the relationship between the passenger and the manufacturing brand via the communication unit. The first calculation unit may calculate the value of the first index so as to increase or decrease according to the reliability and the relationship information acquired by the relationship acquisition unit by referring to the reliability and the relationship information.

[0022] According to the fourth aspect, by referring to the relationship information in addition to the reliability, when calculating the first index, the "strength of connection" of the passenger to the manufacturing brand can be reflected from a different perspective from the reliability. This is advantageous for more appropriately selecting the passengers to be given preferential treatment.

[0023] According to the fifth aspect, the first calculation unit may calculate the value of the first index by referring to at least one of the store visit information of the passenger to the dealership of the moving object, the visit information to the website related to the manufacturing brand, the participation history in the official events related to the manufacturing brand, and the SNS information related to the manufacturing brand as the relationship information.

[0024] According to the fifth aspect, by referring to the store visit information, visit information, participation history, and SNS information, when calculating the value of the first index, the "strength of connection" to the manufacturing brand can be more appropriately reflected. This is advantageous for more appropriately selecting the passengers to be given preferential treatment.

[0025] Further, according to the sixth aspect of the present disclosure, the preferential treatment determination system includes a communication unit connected to the Internet, an influence acquisition unit that acquires influence information characterizing the social influence of the passenger via the communication unit, and a second calculation unit that calculates a second index defined to increase or decrease according to the strength of the social influence based on the influence information acquired by the influence acquisition unit. The preferential treatment determination unit may score the preferential treatment so as to increase or decrease according to both the first index and the second index.

[0026] According to the sixth aspect, in addition to the first index considering the "strength of association", a second index considering the "strength of the passenger's own social influence" is referred to. Thereby, the scoring of the preferential treatment can be performed more appropriately, which is advantageous for more appropriately selecting the passengers to be given preferential treatment.

[0027] Further, according to the seventh aspect of the present disclosure, the second calculation unit calculates the second index by referring to at least one of the exposure history to the mass media and the activity status on the social media as the influence information, and the activity status on the social media includes at least one of the posting history of the passenger whose estimation target is the first index, the number of followers, the number of impressions, and the possibility of monetization on the social media.

[0028] According to the seventh aspect, by referring to the exposure history to the mass media and the activity status on the social media, the "strength of the passenger's own social influence" can be more appropriately reflected when calculating the second index. Thereby, it is advantageous for more appropriately selecting the passengers to be given preferential treatment.

[0029] Furthermore, according to an eighth aspect of this disclosure, the preferential treatment determination system may include a benefit setting unit that sets benefits to be granted to a plurality of passengers according to the level of the preferential treatment value, and the benefit setting unit may grant to the passengers who are designated as preferential treatment at least one of the following: a benefit of lending the mobile vehicle, a benefit of participating in an official event related to the manufacturing brand of the mobile vehicle, a benefit of granting limited functions to the mobile vehicle, and a monetary benefit when purchasing or selling the mobile vehicle.

[0030] According to the eighth aspect described above, by providing benefits according to the degree of preferential treatment, it is possible to encourage PR by passengers selected for preferential treatment and to encourage each passenger to improve their degree of preferential treatment. By encouraging an improvement in the degree of preferential treatment, it is possible to strengthen the connection between each passenger and the mobile vehicle or manufacturing brand, such as by promoting the use of various services, encouraging visits to sales outlets, and strengthening SNS activities.

[0031] Furthermore, according to a ninth aspect of this disclosure, the preferential treatment determination system may include a token granting unit that grants a digital token managed on a blockchain to the passenger designated as eligible for preferential treatment, wherein the value of the digital token fluctuates according to the value of the first indicator or the degree of preferential treatment, and the benefit setting unit sets different benefits to the passenger designated as eligible for preferential treatment according to the value of the digital token.

[0032] According to the ninth embodiment described above, by providing digital tokens to passengers who are designated as eligible for preferential treatment, it is possible to further promote PR by the selected passengers and to improve the level of preferential treatment for each passenger.

[0033] Furthermore, according to a tenth aspect of this disclosure, the candidate selection unit may certify as influencers passengers whose degree of preferential treatment is equal to or greater than a predetermined threshold among the passengers who have been designated as eligible for preferential treatment.

[0034] According to the tenth embodiment described above, by selecting passengers who deserve special treatment as influencers, manufacturers can more appropriately select passengers who should be asked to cooperate in their PR activities.

[0035] Furthermore, according to an eleventh aspect of this disclosure, the preferential treatment determination system may include an interactive terminal mounted on the mobile vehicle, the terminal accepts modifications by the passenger to the content of the service provided, the estimation unit calculates the frequency or amount of modifications to the content of the service provided via the terminal, and the estimation unit updates the reliability score according to the frequency or amount of modifications.

[0036] According to the 11th embodiment described above, for example, if the frequency of revisions is high, it can be assumed that the passenger does not trust the information provided. In that case, it is permissible to underestimate the passenger's level of trust. By referring to the frequency of revisions, the passenger's level of trust can be estimated more appropriately.

[0037] Furthermore, according to a twelfth aspect of this disclosure, the preferential treatment determination system may include an interactive terminal mounted on the mobile vehicle, the terminal receiving whether or not the passenger accepts the content of the service to be provided, the estimation unit calculating the frequency of acceptance of the content to be provided via the terminal, and the estimation unit updating the reliability score according to the frequency of acceptance.

[0038] According to the 12th embodiment described above, for example, when the frequency of adoption is high, it can be assumed that the passenger trusts the content provided. In that case, it is permissible to overestimate the passenger's level of trust. By referring to the frequency of adoption, the passenger's level of trust can be estimated more appropriately.

[0039] Furthermore, according to a thirteenth aspect of this disclosure, the preferential treatment determination system may include an interactive terminal mounted on the mobile vehicle, the terminal receiving the passenger's evaluation of the service provided, the estimation unit aggregating the evaluations received by the terminal, and the estimation unit updating the reliability score according to the aggregated evaluation results.

[0040] According to the 13th embodiment described above, for example, when a passenger gives a high rating to the services provided, it can be assumed that the passenger trusts the services provided. In that case, it is permissible to overestimate the passenger's level of trust. By referring to the ratings, the passenger's level of trust can be estimated more appropriately.

[0041] Furthermore, according to a 14th aspect of this disclosure, the estimation unit may determine the cumulative operating period of the terminal, and the estimation unit may update the reliability according to the cumulative operating period.

[0042] According to the 14th embodiment described above, for example, when the cumulative operating period is long, it can be assumed that the passenger trusts the provided content. In that case, it is permissible to estimate the passenger's level of trust more highly. By referring to the cumulative operating period, the passenger's level of trust can be estimated more appropriately. [Effects of the Invention]

[0043] As explained above, this disclosure allows for a more appropriate selection of passengers who are eligible for preferential treatment. [Brief explanation of the drawing]

[0044] [Figure 1] Figure 1 is a schematic diagram illustrating a preferential treatment determination system and a proposed system used in said preferential treatment determination system. [Figure 2] Figure 2 is a block diagram showing an example configuration of an in-vehicle system. [Figure 3A] Figure 3A is a schematic diagram illustrating the input and output of the proposed system. [Figure 3B] Figure 3B is a schematic diagram illustrating the screen display in the proposed system. [Figure 4] Figure 4 is a schematic diagram illustrating an example of the target of estimation by the estimation unit. [Figure 5] Figure 5 is a diagram illustrating the confidence estimation method. [Figure 6] Figure 6 is a flowchart illustrating the procedure for estimating the level of interest. [Figure 7] Figure 7 is a diagram illustrating the method for estimating uniqueness. [Figure 8] Figure 8 is a flowchart illustrating the procedure for estimating uniqueness. [Figure 9] Figure 9 is a schematic diagram illustrating an example of a control target by the proposed unit. [Figure 10] Figure 10 is a schematic diagram illustrating the control content provided by the proposed unit. [Figure 11] Figure 11 is a block diagram showing an example configuration of a preferential treatment determination device. [Figure 12] Figure 12 is a functional block diagram illustrating the configuration of a preferential treatment determination device. [Figure 13A] Figure 13A is a graph illustrating the relationship between confidence level and the first indicator. [Figure 13B] Figure 13B is a graph illustrating the relationship between flight history and the first indicator. [Figure 13C] Figure 13C is a graph illustrating the relationship between relational information and the first indicator. [Figure 14] Figure 14 is a graph illustrating the relationship between influence information and the second indicator. [Figure 15] Figure 15 is a graph illustrating the relationship between the first indicator, the second indicator, or sales promotion level, and priority. [Figure 16] Figure 16 is a flowchart illustrating the procedure for determining the degree of preferential treatment. [Figure 17] Figure 17 is a diagram illustrating the selection process for preferential customers. [Figure 18] Figure 18 is a diagram illustrating the digital tokens awarded to preferred customers. [Figure 19] Figure 19 is a flowchart illustrating a process based on the degree of preferential treatment. [Modes for carrying out the invention]

[0045] The embodiments of this disclosure will be described below with reference to the drawings. Note that the following description is illustrative.

[0046] [A. Overall composition] Figure 1 is a schematic diagram illustrating the preferential treatment determination system S related to this disclosure and the proposed system 1001 used in the preferential treatment determination system S.

[0047] As shown in Figure 1, the preferential treatment determination system S consists of a proposed system 1001 and a preferential treatment determination device 1002. The proposed system 1001 and the preferential treatment determination device 1002 are electrically connected to each other. The proposed system 1001 is connected to the preferential treatment determination device 1002 so as to be able to transmit data. The preferential treatment determination device 1002 is connected to the Internet Ne2 so as to be able to transmit and receive data via a communication unit 514, which will be described later.

[0048] The preferential treatment determination system S according to this embodiment is a system that analyzes multiple passengers U, U' riding in vehicles V, V', which are moving objects, respectively. This analysis is performed by the preferential treatment determination device 1002. In this analysis, the preferential treatment determination device 1002 utilizes the results of processing performed on each vehicle V by the proposed system 1001, such as the confidence level C1 estimated for each vehicle V.

[0049] The proposed system 1001 will be described in detail below, followed by a detailed explanation of the hardware and software configuration of the preferential treatment determination device 1002, which utilizes the processing performed by the system 1001.

[0050] [B. Proposed System] Figure 2 is a block diagram illustrating an example configuration of the in-vehicle system 1. Figure 3A is a schematic diagram illustrating the input / output in the proposed system 1001. Figure 3B is a schematic diagram illustrating the display screen of terminal 11.

[0051] <1. System Configuration> As illustrated in Figure 1, the proposal system 1001 according to this embodiment is a system that makes proposals to the occupant U of a vehicle V, which is a mobile object. The proposals made by the system 1001 include proposals for occupant actions to be performed by the occupant U themselves, and proposals for mobile object actions that are accepted or rejected by the occupant U and executed by the vehicle V. These proposals are made by a proposal unit 112, which will be described later. This proposal unit 112 is mounted on the vehicle V and corresponds to the "service provision unit" in this embodiment in that it can provide services to the occupant U.

[0052] The occupant's actions include the act of traveling to a predetermined destination Lp. The proposed travel action includes, for example, the proposal of a specific destination Lp and the proposal of a travel route Tp to said destination Lp, as illustrated in Figure 3B. The act of traveling to the destination Lp may be performed by the passenger U driving the vehicle V, or by the passenger U allowing the vehicle V to be driven automatically.

[0053] Here, the term "destination Lp" is used in a broad sense. That is, in this embodiment, destination Lp may be a specific point such as a building, or it may be an entire region including that specific point (so to speak, the target area). Furthermore, the term destination Lp includes events held at a specific location (for example, tourist events).

[0054] The operation of the mobile vehicle includes sound control of vehicle V, lighting control of vehicle V, and air conditioning control of vehicle V. The proposed sound control includes setting the sound equipment 123 and proposing the operation of the sound equipment 123 based on that setting. In addition, the proposed lighting control includes setting the lighting equipment 121 and proposing the operation of the lighting equipment 121 based on that setting. The setting of lighting equipment 121 includes setting the light intensity of the lighting. The proposed air conditioning control includes setting the air conditioning equipment 122 and proposing the operation of the air conditioning equipment 122 based on that setting. The setting of air conditioning equipment 122 includes setting the set temperature, airflow rate, and airflow direction.

[0055] The suggestions for crew actions and vehicle movements are examples of "service provision" in this embodiment. Here, the suggestions for crew actions include the provision of a destination suggestion service configured to suggest the aforementioned travel actions, and, as an alternative example of the suggestions for travel actions, the provision of a tourism support service that provides tourist support information to passenger U. On the other hand, the suggestions for vehicle movements include the provision of an acoustic suggestion service configured to suggest the aforementioned acoustic control.

[0056] Here, the term "passenger U" includes not only the driver of vehicle V but also any passengers in vehicle V. Hereafter, suggestions for passenger actions may be referred to as "action suggestions," and suggestions for vehicle movement may be referred to as "movement suggestions." Details of action suggestions and movement suggestions will be described later.

[0057] Furthermore, the vehicle V, as a mobile entity, may be an automobile using an internal combustion engine as its power unit, an electric vehicle using an electric motor, or a hybrid vehicle using both an internal combustion engine and an electric motor. The fuel used in the internal combustion engine may include any fuel, such as gasoline, diesel fuel, hydrogen gas, and other gaseous fuels. In this embodiment, vehicle V is a four-wheel drive vehicle. Vehicle V is an SDV (Software Defined Vehicle) and can provide various functions to the passenger U, such as the services described later.

[0058] As shown in Figure 1, the proposed system 1001 comprises an in-vehicle system 1 mounted on a vehicle V, and an external server 300 connected to the in-vehicle system 1 via a cloud network Ne1. It is not mandatory to consider the external server 300 as an element of the proposed system 1001. The external server 300 may also be considered as an element of the preferential treatment determination device 1002.

[0059] Here, as shown in Figure 2, the in-vehicle system 1 includes an in-vehicle device 10, an in-vehicle computer 100, and an in-vehicle database 200. The in-vehicle device 10 is electrically connected to the in-vehicle computer 100 and the in-vehicle database 200 via an in-vehicle network.

[0060] The in-vehicle device 10 includes a terminal 11, a vehicle control device 12, a biometric information detection device 13, and a movement information detection device 14. All of these devices are mounted on the vehicle V.

[0061] (1-1. Terminal) Terminal 11 notifies passenger U of the proposal from the proposal unit 112. More specifically, terminal 11 has both an action suggestion notification function and an action suggestion notification function. Furthermore, terminal 11 also has a function to present questionnaires to passenger U.

[0062] Specifically, the terminal 11 according to this embodiment is composed of a display such as a liquid crystal panel or an organic EL panel, and a speaker. This terminal 11 can also be considered a notification unit that notifies the passenger U by visualizing the proposed content from the proposal unit 112 and displaying it on the display screen Scr. The terminal 11 can also be considered a notification unit configured to output the proposed content from the proposal unit 112 as sound.

[0063] It is not mandatory to configure terminal 11 with both a display and a speaker. Terminal 11 may be configured with a display alone, or with a speaker alone.

[0064] For example, terminal 11 may display text indicating the proposed content, or locations or routes superimposed on a map, on the display screen Scr, or emit sound corresponding to the text. In addition to such displays or sounds, terminal 11 can also visualize and display the survey response items using text or the like.

[0065] Terminal 11 is also configured as an interactive terminal device. Specifically, terminal 11 is configured to accept or reject a proposal from passenger U, to receive modifications from passenger U to the proposal, and to receive the passenger's evaluation of the proposal. Here, the term "proposal" can be replaced with "service provision content" or simply "service content."

[0066] More specifically, terminal 11 can also receive responses to a first questionnaire A1 that characterizes the passenger U's personal characteristics C0, particularly the preference characteristics C7 described later, and responses to a second questionnaire A2 that characterizes the passenger U's purpose of boarding C4 (see Figure 4 below).

[0067] Specifically, the terminal 11 according to this embodiment is configured to use a touch panel on the display as described above. This terminal 11 can also be considered as a reception unit that receives input and operations performed by the passenger.

[0068] It is not mandatory to configure terminal 11 using a touch panel. The reception section of terminal 11 may be configured by providing a separate microphone instead of, or in addition to, a touch panel.

[0069] Terminal 11 then receives responses regarding passenger U's age, gender, hobbies, height, weight, body type, and family structure as the first questionnaire A1. Note that including age and gender in the response items is not mandatory. As described later, depending on the usage flag, questionnaires regarding age, gender, height, weight, and body type may be omitted.

[0070] Terminal 11 also accepts responses as the second questionnaire A2, relating to the passenger U's destination Lp and the purpose of travel to that destination Lp. Specifically, the second questionnaire A2 includes questions that can determine whether or not the purpose of travel is to a particular destination Lp.

[0071] If the purpose is to travel to a specific destination Lp, terminal 11 will accept a response regarding the purpose of travel to that destination Lp as the second questionnaire A2. If the purpose is not to travel to a specific destination Lp, terminal 11 may accept a response regarding the region that passenger U considers to be the destination Lp (destination region).

[0072] The content I1 of the conversation between passenger U and terminal 11 is transmitted to the in-vehicle database 200 as electronic data corresponding to the content I1. The in-vehicle database 200 stores the content I1 (see Figures 2 and 4). This content I1 includes responses to the first questionnaire A1 and the second questionnaire A2, as well as the acceptance, revision, and evaluation of proposals made by passenger U. The content I1 may be stored in the storage 103 of the in-vehicle computer 100 instead of, or in addition to, storage in the in-vehicle database 200.

[0073] (1-2. Vehicle control devices) The vehicle control device 12 includes lighting equipment 121, air conditioning equipment 122, and sound equipment 123. Each of these devices functions as an actuator that controls the vehicle V by receiving control signals from the onboard computer 100.

[0074] Lighting equipment 121 functions as lighting to illuminate the interior of the vehicle. Air conditioning equipment 122 performs air conditioning such as heating, cooling, and ventilation to maintain the environment inside the vehicle. Audio equipment 123 controls the sound inside the vehicle. For example, audio equipment 123 can play music inside the vehicle or provide voice notifications of various suggestions to passenger U. Voice notifications of suggestions are not mandatory.

[0075] (1-3. Biometric Information Detection Devices) The biometric information detection device 13 includes an in-vehicle camera 131 and an in-vehicle microphone 132. These devices function as sensing devices that detect the biometric information I2 of the passenger U. The biometric information detection device 13 is an example of the "biometric information detection unit" in this embodiment.

[0076] The biometric information I2 includes a body image of the passenger U acquired via the in-vehicle camera 131, which acts as an imaging unit, and voice data of the passenger U acquired via the in-vehicle microphone 132.

[0077] The in-vehicle camera 131 captures images of the interior of the vehicle. More specifically, the in-vehicle camera 131 captures images of the body of the occupant U of the vehicle V. More specifically, the in-vehicle camera 131 captures one or more of the occupant U's facial expression, clothing, gaze, blinking, posture, and body movements. The in-vehicle camera 131 operates by receiving control signals from the in-vehicle computer 100 and generates electrical signals indicating the captured images.

[0078] The in-vehicle microphone 132 collects sound from inside the vehicle. For example, the in-vehicle microphone 17 collects conversations taking place inside the vehicle. The in-vehicle microphone 132 operates by receiving control signals from the in-vehicle computer 100 and generates electrical signals indicating the sound it has collected.

[0079] The electrical signals generated by each sensing device constituting the biometric information detection device 13 are transmitted to the in-vehicle database 200 as electronic data representing the biometric information I2 of the passenger U. The in-vehicle database 200 stores the biometric information I2 (see Figure 2). The biometric information I2 may be stored in the in-vehicle computer 100 instead of, or in addition to, storage in the in-vehicle database 200.

[0080] In addition, the biometric information detection device 13 may include a heart rate sensor for detecting the passenger U's heart rate, an electromyography sensor for detecting the passenger U's electromyographic potential, an electroencephalogram (EEG) sensor for detecting the passenger U's brain waves, a temperature sensor for detecting the passenger U's body temperature, and the like.

[0081] Furthermore, the biometric information I2 includes, in addition to the information detected by the sensing device described above, information contained in the dialogue content I1 via the terminal 11, or information that can be inferred from the dialogue content I1. The biometric information I2 based on the dialogue content I1 includes physical information of the passenger U, such as height, weight, body type, and gender, and information indicating the passenger U's physical condition (the passenger U's own subjective judgment).

[0082] The physical information and health condition of passenger U may be determined based on the images captured by the in-vehicle camera 131. In this case, passenger U's health condition will be determined by objective information such as passenger U's facial color, rather than by information based on passenger U's own subjective judgment.

[0083] (1-4. Movement Information Detection Device) The movement information detection device 14 includes an external camera 141 and a driving environment sensor 142. These devices function as sensing devices that detect movement information I3 that characterizes the characteristics of the vehicle V itself and the driving conditions of the vehicle V. The movement information detection device 14 is an example of the "movement information detection unit" and "mounted equipment for the moving body" in this embodiment.

[0084] The external camera 141 captures images of the area outside the vehicle. More specifically, the external camera 141 captures images of the surrounding environment of vehicle V. Here, the term "surrounding environment of vehicle V" includes preceding vehicles, following vehicles, oncoming vehicles, pedestrians, various structures, and the traffic environment of vehicle V. The external camera 141 operates by receiving control signals from the onboard computer 100 and generates electrical signals indicating the captured images.

[0085] The driving environment sensor 142 detects the weather outside the vehicle, the location information of the vehicle V, and the surrounding environment. The driving environment sensor 142 consists of an illuminance sensor, a raindrop sensor, a GPS sensor, a LiDAR (Light Detection And Ranging) sensor, an outside temperature sensor, and a sound-collecting microphone. The illuminance sensor detects the illuminance of sunlight. The raindrop sensor detects raindrops caused by rainfall. The GPS sensor detects the location information of the vehicle V. The LiDAR emits laser light and detects the distance to an object or the shape of an object based on the reflected light. A radar that emits radio waves or millimeter waves may be used instead of LiDAR. Each of these devices generates an electrical signal indicating the detection result. The outside temperature sensor detects the temperature outside the vehicle. The sound-collecting microphone detects the presence and volume of noise outside the vehicle.

[0086] The electrical signals generated by each sensing device constituting the movement information detection device 14 are transmitted to the in-vehicle database 200 as electronic data indicating the movement information of the vehicle V. The in-vehicle database 200 stores the movement information I3 (see Figure 2). The movement information I3 may be stored in the in-vehicle computer 100 instead of, or in addition to, storage in the in-vehicle database 200.

[0087] In addition, the mobility information detection device 14 may include a fuel level sensor for detecting the remaining amount of fuel or battery, a history sensor for detecting the passenger U's riding history, a vehicle speed sensor for detecting the vehicle speed of the vehicle V, an accelerator pedal position sensor for detecting the degree of accelerator pedal opening, and the like.

[0088] Furthermore, the movement information I3 includes, in addition to the information detected by the sensing device described above, information contained in the dialogue content I1 via the terminal 11 or information estimated from the dialogue content I1. The movement information I3 obtained or estimated from the dialogue content I1 includes information indicating the purpose of boarding C4, such as the destination specified by the passenger U.

[0089] Furthermore, the mobility information I3 includes a usage flag indicating whether the vehicle V, as a mobile entity, is a shared item (a so-called "shared car") or an item temporarily leased (a so-called "rental car"). This usage flag is mobility information I3 that characterizes the characteristics of the vehicle V itself and is stored in advance in the in-vehicle database 200 or the in-vehicle computer 100.

[0090] Furthermore, the movement information I3 includes a time flag indicating the date and time of use of the vehicle V as a moving object. This time flag is movement information that characterizes the driving status of the vehicle V and is counted by a timer 105 provided in or connected to the onboard computer 100.

[0091] (1-5. In-vehicle database) The in-vehicle database 200 stores the content I1 of the conversation between terminal 11 and passenger U, passenger U's biometric information I2, and the movement information I3 of passenger U and vehicle V. The in-vehicle database 200 is composed of storage capable of continuously storing various types of information. The in-vehicle database 200 can be composed of storage based on any standard, such as storage employing flash memory.

[0092] The in-vehicle database 200 is located inside the vehicle V and stores the content of the conversation between the passenger U and the terminal 11. Together with the external server 300 described later, it constitutes the "storage unit" according to this embodiment.

[0093] (1-6. Onboard Computer) The in-vehicle computer 100 is a computer in which a control unit 101 and an input / output interface (I / O interface) 102 are interconnected via a system bus.

[0094] The control unit 101 includes a processor 101a that executes various programs, a read-only ROM 101b that stores the BIOS and the like, and RAM 101c that functions as main memory.

[0095] In addition to the aforementioned terminal 11, vehicle control device 12, biometric information detection device 13, movement information detection device 14, timer 105, and in-vehicle database 200, the input / output interface 102 is also connected to the storage 103 and the communication control unit 104. The connection via the input / output interface 102 may also be a connection via an in-vehicle network such as CAN.

[0096] The storage 103 stores various information related to the vehicle V and the passenger U, and various programs for processing said information. The in-vehicle database 200 may also serve as the storage 103.

[0097] The communication control unit 104 is connected to the external server 300 via the cloud network Ne1 based on a predetermined communication standard. The in-vehicle computer 100 sends and receives data to and from the external server 300 via the communication control unit 104 and the cloud network Ne1.

[0098] Here, the external server 300 stores a destination database 301 associated with the passenger U's personal characteristics C0, an acoustic database 302 corresponding to the passenger U's personal characteristics C0, and a tourism database 303. The tourism database 303 consists of a database of regional information including event information and tourism information for each destination Lp. The passenger U's personal characteristics C0 will be described later.

[0099] The external server 300 also stores the content I1 of conversations I1 between the terminal 11 of a vehicle V' other than the proposed vehicle V and a group of other users U' other than the passenger U, biometric information I2 of the group of other users U', and movement information I3 of the group of other users U' and the other vehicle V'.

[0100] The external server 300 is located outside the vehicle V' and stores the content of conversations between other user groups U' and other terminals 11. Together with the aforementioned in-vehicle database 200, it constitutes the "storage unit" according to this embodiment.

[0101] As illustrated in Figure 3A, the in-vehicle computer 100 according to this embodiment comprises an estimation unit 111, a proposal unit 112, a terminal control unit 113, and an operation control unit 114. The proposal unit 112 is composed of an action proposal unit 112A and an operation proposal unit 112B. Details of these elements will be described below. These elements are, for example, composed of the control unit 101 of the in-vehicle computer 100.

[0102] <2. Estimation part> Figure 4 is a schematic diagram showing the target of estimation by the estimation unit 111. Figure 5 is a diagram for explaining the estimation procedure for confidence level C1. Figure 6 is a flowchart illustrating the estimation procedure for interest level C2. Figure 7 is a diagram for explaining the estimation procedure for originality level C3. Figure 8 is a flowchart illustrating the estimation procedure for originality level C3.

[0103] The estimation unit 111 estimates the passenger U's personal characteristics C0, the passenger U's purpose for boarding C4, the passenger U's mental and physical state including emotions C5, and the passenger U's boarding risk C6. Note that estimating the level of interest C2 in personal characteristics C0, and estimating the purpose for boarding C4, mental and physical state C5, and boarding risk C6 are not mandatory. Personal characteristics C0 is an example of "personal characteristics" in this embodiment.

[0104] Specifically, the estimation unit 111 according to this embodiment is composed of a machine learning model. The machine learning model that constitutes the estimation unit 111 may be a neural network or a rule-based model. It is not essential that the estimation unit 111 be composed of a machine learning model.

[0105] Specifically, the estimation unit 111 estimates personal characteristics C0, purpose of boarding C4, physical and mental state C5, and boarding risk C6 based on at least one of the following: the content of the conversation with the passenger U via the terminal 11, biometric information detected by the biometric information detection device 13, and movement information detected by the movement information detection device 14. The conversation content, biometric information, and movement information are read from the in-vehicle database 200, respectively.

[0106] Here, personal characteristics C0 include judgment characteristics Ca, which characterize the passenger U's judgment criteria for the proposal from proposal unit 112, and preference characteristics C7, which characterize the passenger U's preferences that may influence whether or not the proposal from proposal unit 112 is accepted. Note that preference characteristics C7 are not essential in this disclosure.

[0107] The judgment characteristic Ca, as illustrated in Figure 4, includes the level of trust C1 between the passenger U and the proposal unit 112, the passenger U's level of interest C2 in the proposal from the proposal unit 112, and the passenger U's originality C3 in deciding whether to accept or reject the proposal from the proposal unit 112. On the other hand, the preference characteristic C7 is related to the passenger U's life stage and lifestyle.

[0108] The judgment characteristic Ca according to this embodiment is composed of a combination of confidence level C1, interest level C2, and uniqueness level C3. Note that interest level C2 and uniqueness level C3 are not essential in this disclosure.

[0109] (2-1. Estimation of confidence level) The estimation unit 111 estimates the confidence level C1 of passenger U based on the passenger U's response to the proposal (offer) from the proposal unit 112, which acts as the service provider. The estimation based on the response can be performed based on the dialogue content I1.

[0110] The confidence level C1 estimated by the estimation unit 111 represents the level of confidence between the passenger U and the vehicle V as a mobile entity, the services provided by the proposed system 1001, or the manufacturer brand of the vehicle V.

[0111] The term "services provided by the proposed system 1001" used here is used in a broad sense. In other words, the term "confidence level C1 between passenger U and the services provided by the proposed system 1001" includes both passenger U's confidence level C1 in the proposed content (contents provided, services provided) from the proposed system 1001 and passenger U's confidence level C1 in the proposed system 1001 itself.

[0112] Furthermore, the term "manufacturing brand" used here is used in a broad sense. The phrase "trust level C1 between passenger U and the manufacturing brand of vehicle V" includes both passenger U's trust level C1 in the manufacturer of vehicle V and passenger U's trust level C1 in the manufacturing brand.

[0113] Here, the estimation unit 111 calculates the degree of update of the individual characteristic C0 (for example, the number of updates). Then, the estimation unit 111 updates the confidence level C1 according to the calculated degree of update. Specifically, as illustrated in graph G1 of Figure 5, the estimation unit 111 sets a higher confidence level C1 the more updates are calculated.

[0114] Furthermore, the estimation unit 111 calculates the frequency or amount of modification of the proposal (content provided) via the terminal 11. The estimation unit 111 then updates the confidence level C1 according to the calculated frequency or amount of modification. Specifically, as illustrated in graph G2 of Figure 5, the estimation unit 111 sets the confidence level C1 lower the higher the calculated frequency or the larger the amount of modification.

[0115] Furthermore, the estimation unit 111 calculates the frequency or number of times a proposal (content offered) is adopted via the terminal 11. The estimation unit 111 then updates the confidence level C1 according to the calculated frequency or number of adoptions. Specifically, as illustrated in graph G3 of Figure 5, the estimation unit 111 sets a higher confidence level C1 the higher the calculated frequency or number of adoptions.

[0116] Furthermore, the estimation unit 111 aggregates the evaluations of the proposals (offerings) received by the terminal 11. The estimation unit 111 then updates the confidence level C1 according to the aggregated evaluation results. Specifically, as illustrated in graph G4 of Figure 5, the estimation unit 111 sets a higher confidence level C1 the more highly the proposal is evaluated.

[0117] Furthermore, the estimation unit 111 calculates the cumulative operating period (valid period) since the terminal 11 activated the proposed function (service provision function). The estimation unit 111 then updates the confidence level C1 according to the length of the valid period. Specifically, as illustrated in graph G5 of Figure 5, the estimation unit 111 sets the confidence level C1 higher as the valid period lengthens.

[0118] Thus, the estimation unit 111 according to this embodiment calculates the confidence level C1 using five variables that characterize the passenger U's response to the proposal from the proposal unit 112: a first variable that quantifies the degree of update of personal characteristics C0, a second variable that quantifies the frequency or amount of revision of the proposal, a third variable that quantifies the frequency or number of times the proposal is adopted, a fourth variable that quantifies the evaluation of the proposal, and a fifth variable that quantifies the cumulative operating period of the terminal 11.

[0119] In detail, a machine learning model is pre-stored in a storage device such as storage 103, which takes the five variables as input and outputs a confidence score C1 value. The estimation unit 111 calculates the confidence score C1 value by inputting the five variables into the machine learning model. As mentioned above, using a machine learning model is not mandatory.

[0120] The proposal unit 112 then executes a proposal (service provision) as described below, based on the confidence level C1 calculated (updated) in this manner. The estimation unit 111 further updates the confidence level C1 of passenger U based on the passenger U's response to the proposal.

[0121] In this way, the proposal unit 112 and the estimation unit 111 perform feedback control via the confidence level C1. Through this feedback control, the proposal from the proposal unit 112 is optimized according to the confidence level C1 of the passenger U.

[0122] (2-2. Estimation of the level of interest) The estimation unit 111 estimates the passenger U's level of interest C2 based on the passenger U's response to the proposal from the proposal unit 112. This level of interest C2 may be considered as the level of interest in the proposal unit 112, the proposal system 1001, or the proposed content (content offered, services offered) from the proposal system 1001.

[0123] Specifically, the estimation unit 111 according to this embodiment determines the passenger U's level of interest C2 based on the passenger U's behavioral change stages, which are represented by multiple stages. Here, the behavioral change stages are modeled by five stages, as is well known. The five stages include the "precontemplation stage," the "contemplation stage," the "preparation stage," the "action stage," and the "maintenance stage."

[0124] The "precontemplation stage" is the stage in which a person does not intend to change their "behavior" within a predetermined first period (e.g., six months). The "contemplation stage" is the stage in which a person intends to change their "behavior" within a predetermined third period (e.g., one month). The "action stage" is the stage in which a person intends to change their "behavior" within a second period shorter than the aforementioned first period (e.g., one month). The "action stage" is the stage in which a person has changed their "behavior" but has not yet reached the predetermined third period (e.g., six months). The "maintenance stage" is the stage in which a person has changed their "behavior" and has reached the aforementioned third period or longer.

[0125] In this embodiment, "action" includes "adoption of a proposal from the proposed system 1001." Furthermore, this embodiment utilizes four stages, excluding the "preparation phase."

[0126] More specifically, the estimation unit 111 is configured to determine the progress of the dialogue between terminal 11 and passenger U, and to update the level of interest C2 according to that progress. More specifically, the estimation unit 111 updates the level of interest C2 by referring to the confidence level C1 in addition to the progress.

[0127] As an example, as illustrated in step S101 of Figure 6, the estimation unit 111 first determines whether the proposal function of the proposal system 1001 (specifically, the proposal function for action proposals or movement proposals) is enabled. If the determination in step S101 is NO, the estimation unit 111 sets the interest level C2 of passenger U to "precontemplation" in step S102 and returns the control process to step S101. The estimation unit 111 maintains the interest level C2 at "precontemplation" until the proposal function is enabled.

[0128] If the determination in step S101 is YES, the estimation unit 111 proceeds to step S103 of the control process. In step S103, the estimation unit 111 determines whether there is a history of acquiring personal characteristics C0 or whether the acquisition of new personal characteristics C0 has been completed. If the determination in step S103 is NO, the estimation unit 111 sets the interest level C2 of passenger U to "interest stage" in step S104 and returns the control process to step S103. If there is no history of acquiring personal characteristics C0, the estimation unit 111 keeps the interest level C2 at "interest stage" until the acquisition of new personal characteristics C0 is completed.

[0129] If the determination in step S103 is YES, the estimation unit 111 proceeds to step S105 of the control process. In step S105, the estimation unit 111 determines whether the elapsed time since the proposed function was activated (system validity period) has exceeded a predetermined first period, or whether the number of proposed options calculated based on the confidence level C1 has exceeded a predetermined first threshold. If the determination in step S105 is NO, the estimation unit 111 sets the interest level C2 of passenger U to "execution period" in step S106 and returns the control process to step S105. The estimation unit 111 maintains the interest level C2 at "execution period" until the system validity period exceeds the first period or the number of proposed options exceeds the first threshold.

[0130] Here, the system validity period is an example of the "cumulative operating period" in this embodiment. Furthermore, updating the level of interest C2 by referring to the number of proposed options is substantially equivalent to updating the level of interest C2 by referring to the confidence level C1, as will be described later.

[0131] If the determination in step S105 is YES, the estimation unit 111 proceeds to step S107 of the control process. In step S107, the estimation unit 111 determines whether the proposed function has been disabled. If the determination in step S107 is NO, the estimation unit 111 sets the interest level C2 of passenger U to the "maintenance phase" in step S108 and returns the control process to step S107. The estimation unit 111 maintains the interest level C2 in the "maintenance phase" until the proposed function is disabled.

[0132] If the determination in step S107 is YES, the estimation unit 111 sets the interest level C2 of passenger U to "incontemplation" in step S109 and terminates the control process in Figure 6. In this case, with the interest level C2 set to "incontemplation," the estimation unit 111 will execute the flow in Figure 6 again, starting from step S101.

[0133] (2-3. Estimation of Uniqueness) The estimation unit 111 quantifies and estimates the uniqueness C3 of passenger U in deciding whether to accept or reject a proposal from the proposal unit 112, based on the content I1 of the dialogue between passenger U and other user group U' with terminal 11. This uniqueness C3 indicates the uniqueness of passenger U's decision to accept or reject a proposal compared to the decisions of other user groups. This uniqueness C3 may also be considered as the strength of subjectivity in passenger U's decision to accept or reject a proposal.

[0134] More specifically, the estimation unit 111 estimates the uniqueness C3 of the passenger U based on the contents of the in-vehicle database 200 and the external server 300, which are located inside and outside the vehicle V and serve as memory units.

[0135] First, in step S201 of Figure 8, the estimation unit 111 reads the response results of other user group U' to the first questionnaire A1 from the external server 300 as the content I1 of the dialogue with terminal 11.

[0136] Next, in step S202, the estimation unit 111 performs statistical processing on the information read from the external server 300 and creates a distribution diagram M1 that represents the statistical distribution of the response results of other user groups U'. As illustrated in Figure 7, the distribution diagram M1 is represented by a distribution function such as a normal distribution or a Pareto distribution.

[0137] Next, in step S203, the estimation unit 111 reads the passenger U's response result Vu to the first questionnaire A1 from the in-vehicle database 200 as the content I1 of the conversation with the terminal 11.

[0138] Next, in step S204, the estimation unit 111 superimposes the passenger U's response result Vu onto the distribution diagram M1 and quantifies the passenger U's uniqueness C3 based on the superimposed result. For example, if the distribution diagram M1 is represented by a normal distribution, the difference ΔV between the user mean Va on that normal distribution and the response result Vu can be used for uniqueness C3. As an example, as shown in Figures 7(a) and 7(b), uniqueness C3 may be quantified by using parameters, coefficients, variables, functions, etc., such that the difference ΔV is low when the difference ΔV is small and high when the difference ΔV is large.

[0139] (2-4. Estimation of Preference Characteristics) The estimation unit 111 estimates the passenger U's preference characteristics C7 based on the stored contents of the in-vehicle database 200 (particularly the dialogue content I1) (see Figure 4). The preference characteristics C7 include an index representing the passenger U's age in multiple stages, an index representing the passenger U's gender, an index representing the passenger U's hobbies, an index representing the passenger U's family structure, an index representing other characteristics of the passenger U, and an index representing the relationship between the passenger U and the vehicle V. The values ​​of these indexes may be quantified.

[0140] The age of passenger U can be expressed in six stages, for example, "infancy," "childhood," "adolescence," "adulthood," "middle age," and "old age." The family structure of passenger U can be expressed in multiple stages, for example, "single," "nuclear family," and "multi-generational family."

[0141] Other characteristics of passenger U are represented by one or more flags, such as whether passenger U is a student or whether passenger U is pregnant or raising children.

[0142] An index representing the relationship between passenger U and vehicle V consists of, for example, one or more of the following: frequency of riding vehicle V, duration of riding vehicle V, distance traveled by vehicle V, and location information of vehicle V. These indexes may be further classified by day of the week, time of day, day of the week, or season. These indexes can be calculated based on travel information I3.

[0143] The estimation unit 111 estimates the passenger U's preference characteristics C7 based on at least one of the following: the content of the conversation with the terminal 11 I1, biometric information I2, and mobility information I3. The information referenced may differ for each type of preference characteristic C7. For example, the passenger U's age can be estimated based on the response to the first questionnaire A1. On the other hand, an index representing the relationship between passenger U and vehicle V can be estimated, for example, based on mobility information I3 as described above.

[0144] The passenger U's preference characteristics C7 can also be estimated based on the passenger U's body image. For example, the estimation unit 111 extracts the passenger U's face image based on the passenger U's body image. The estimation unit 111 then estimates the passenger U's gender, age, etc., based on the passenger U's face image. In other words, the preference characteristics C7 can be estimated based on the biometric information I2.

[0145] (2-5. Estimation of the purpose of boarding) The estimation unit 111 obtains the purpose of boarding the vehicle V C4 based on at least one of the following: dialogue content I1, biometric information I2, and movement information I3 (see Figure 4). The estimation unit 111 classifies the purpose of boarding C4 into one of several categories. These categories include one or more of the following: sightseeing, driving, visiting family, funeral, celebration, business, daily life, and other purposes. The degree of each of these categories may also be quantified.

[0146] More specifically, the estimation unit 111 refers to the response to the second questionnaire A2 as dialogue content I1, and estimates the passenger U's purpose of boarding C4 based on that response. For example, if the response to the second questionnaire A2 includes the passenger U's purpose of travel to a specific tourist destination, the estimation unit 111 classifies the purpose of boarding C4 as tourism. On the other hand, if the response to the second questionnaire A2 includes the passenger U's purpose of travel to a specific wedding venue, the estimation unit 111 classifies the purpose of boarding C4 as a celebratory event.

[0147] In addition to, or instead of, referring to the responses to the second questionnaire A2, the estimation unit 111 can estimate the purpose of boarding C4 of passenger U based on the passenger U's body image as biometric information I2. For example, the estimation unit 111 estimates the passenger U's clothing based on the passenger U's body image. The estimation unit 111 then performs a classification of the purpose of boarding C4 according to the estimated clothing.

[0148] To give a further example, if passenger U is wearing a business suit, the estimation unit 111 can classify the purpose of boarding C4 as business. If passenger U is wearing a tie, the estimation unit 111 can classify the purpose of boarding C4 as business, funeral, or celebration based on the pattern and color of the tie.

[0149] Furthermore, in addition to, or instead of, referring to the responses to the second questionnaire A2, the estimation unit 111 can estimate the passenger U's purpose of boarding C4 based on the passenger U's voice data as biometric information I2. This estimation can be performed in combination with estimation based on body images.

[0150] In addition, the estimation unit 111 can estimate the purpose of boarding C4 based on the time flag of the vehicle V. For example, if the vehicle V is used on a weekday, the estimation unit 111 determines that the purpose of boarding C4 is likely to be business or daily life, and if the vehicle V is used on a Saturday or Sunday, it determines that the purpose of boarding C4 is likely to be for a purpose other than business or daily life.

[0151] The estimation unit 111 can improve the accuracy of estimating the purpose of boarding C4 by combining the judgment based on the time flag with the judgment based on the second questionnaire A2, body image, and voice data.

[0152] In addition, the estimation unit 111 can estimate the purpose of boarding C4 based on the vehicle V's usage flag. For example, if the vehicle V is a rental car, the estimation unit 111 can determine that the purpose of boarding C4 is likely to be for a purpose other than business or daily life.

[0153] The estimation unit 111 can improve the accuracy of estimating the purpose of boarding C4 by combining the judgment based on the purpose flag with the judgment based on the second questionnaire A2, body image, voice data, and time flag.

[0154] The usage flag can be used in the response items of the first questionnaire A1. The estimation unit 111 determines whether vehicle V is a shared car or a rental car based on the usage flag. If it is determined to be a shared car or a rental car, the estimation unit 111 reduces the number of response items in the first questionnaire A1 compared to when it is not determined to be a shared car or a rental car. More specifically, if it is determined that vehicle V is a shared car or a rental car, the estimation unit 111 reduces the number of response items other than the usage flag in the first questionnaire A1 compared to when it is not determined to be a shared car or a rental car.

[0155] (2-6. Estimation of physical and mental state) The estimation unit 111 estimates the mental and physical state C5 of the passenger U based on at least one of the following: dialogue content I1, biometric information I2, and movement information I3 (see Figure 4). The mental and physical state C5 estimated by the estimation unit 111 may include the passenger U's physical condition in addition to the passenger U's emotions as described above. As illustrated in Figure 4, the passenger U's physical condition may be the passenger U's physical condition. The estimation unit 111 classifies the passenger U's emotions (and physical condition) into one of several categories. The several categories include one or more of the following: "normal," "joy (happiness)," "anger," "sadness," "fun," "pleasant," "unpleasant," "active (awakened)," "inactive (sleep)," "excited," "calm," "depressed," "relieved," "anxious," "fulfilled," "empty," "disappointed," "fatigued," "weariness," and "exhausted." The degree of each of these categories may also be quantified.

[0156] More specifically, the estimation unit 111 estimates the mental and physical state C5 of passenger U based on the passenger U's body image as biometric information I2. For example, the estimation unit 111 extracts a face image of passenger U based on the passenger U's body image. The estimation unit 111 then estimates the mental and physical state C5 of passenger U based on the facial expression shown in the passenger U's face image.

[0157] Furthermore, in addition to, or instead of, the body image of the passenger U, the estimation unit 111 can estimate the passenger U's mental and physical state C5 based on the passenger U's voice data as biometric information I2. This estimation can be performed in combination with estimation based on the body image. Moreover, the estimation unit 111 can also estimate the passenger U's mental and physical state C5 based on other biometric information I2, such as heart rate.

[0158] In addition, the estimation unit 111 can estimate the emotions of passengers U based on the purpose C4 of boarding the vehicle V. For example, if the purpose of boarding is a funeral, the estimation unit 111 can determine that the emotion is likely to be classified as "sadness." Also, if the purpose of boarding is sightseeing, the estimation unit 111 can determine that the emotion is likely to be classified as "joy" or "happiness."

[0159] The estimation unit 111 can improve the accuracy of estimating the mental and physical state C5 by combining the judgment based on the purpose of boarding C4 with the judgment based on physical images and voice data.

[0160] In addition, the estimation unit 111 can estimate the mental and physical state C5 based on the time flag of the vehicle V. For example, if the vehicle V is used in the early morning or late at night, the estimation unit 111 will determine that there is a high probability that the passenger U is feeling sleepy.

[0161] The estimation unit 111 can improve the accuracy of estimating the mental and physical state C5 by combining the judgment based on the time flag with the judgment based on the physical image, voice data, and purpose of boarding C4.

[0162] (2-7. Estimation of boarding risks) The estimation unit 111 obtains the boarding risk C6 for vehicle V based on at least one of the following: dialogue content I1, biometric information I2, and movement information I3. This boarding risk C6 is represented by two variables: safety risk and economic risk. The boarding risk C6 increases as the safety risk increases and as the economic risk increases.

[0163] More specifically, the estimation unit 111 estimates the boarding risk C6 based on the boarding purpose C4 estimated by the estimation unit 111. For example, the estimation unit 111 collects news of incidents that occurred around the destination Lp of the vehicle V or on the travel route Tp to the destination Lp, and evaluates the safety risk corresponding to the boarding purpose C4 based on the collected results.

[0164] To illustrate further, the estimation unit 111 evaluates the economic risk corresponding to the purpose of travel C4 based on the travel distance and route to the destination Lp. For example, the estimation unit 111 evaluates the economic risk higher the longer the travel distance to the destination Lp.

[0165] In addition, the estimation unit 111 can estimate the boarding risk C6 based on the physical and mental condition C5 of the vehicle V. For example, if the passenger U is unwell, the estimation unit 111 will evaluate the safety risk as higher.

[0166] <3. Action proposal department> Figure 9 is a schematic diagram illustrating an example of a control target by the proposal unit 112. Figure 10 is a schematic diagram illustrating an example of the control content by the proposal unit 112. Figure 9 is common to both the action proposal unit 112A and the operation proposal unit 112B in the proposal unit 112. In Figure 10, underlined items indicate processing specific to either the action proposal unit 112A or the operation proposal unit 112B, while ununderlined items indicate processing common to both.

[0167] The action suggestion unit 112A according to this embodiment is composed of a machine learning model. The machine learning model that constitutes the action suggestion unit 112A may be a neural network or a rule-based model. It is not mandatory to constitute the action suggestion unit 112A with a machine learning model.

[0168] (3-1.Basic concept) The action suggestion unit 112A proposes the crew actions defined above to the passenger U. The action suggestion unit 112A proposes to the passenger U the act of moving to a predetermined destination Lp as the crew action. Specifically, the action suggestion unit 112A performs at least one of the following as the act of moving to a predetermined destination Lp: proposing the destination Lp and proposing the travel route Tp to said destination Lp.

[0169] The action suggestion unit 112A makes a travel suggestion based on the response to the second questionnaire A2 or the boarding purpose C4 estimated by the estimation unit 111 itself. In doing so, the action suggestion unit 112A proposes one or more travel actions.

[0170] For example, if the destination Lp, such as a specific building, is clearly known, the action suggestion unit 112A will suggest a travel route Tp to that destination Lp. As mentioned above, the number of travel route Tp suggestions will be one or more.

[0171] Furthermore, if the destination Lp is only known to the region or municipality, and there is room for suggesting specific destinations, the action suggestion unit 112A will suggest details of the destination (destination Lp), such as a specific building or the location of a tourist event. As mentioned above, the number of destination Lp suggestions will be one or more.

[0172] More specifically, as illustrated in Figure 9, the action suggestion unit 112A determines the crew actions to be suggested to passenger U based on the preference characteristics C7 estimated from the results of the responses to the second questionnaire A2. Further details show that when determining the crew actions to be suggested to passenger U, the action suggestion unit 112A refers to the aforementioned destination database 301 stored in the external server 300. In this destination database 301, as described above, travel activities corresponding to preference characteristics C7 are databased.

[0173] The crew action determined by the action suggestion unit 112A is proposed to the passenger U via the terminal 11. In this embodiment, the action suggestion unit 112A controls the proposal of crew actions based on the confidence level C1 and interest level C2 of the individual characteristics C0.

[0174] In detail, the action suggestion unit 112A controls the suggestion of crew actions by referring to originality C3, purpose of boarding C4, physical and mental state C5, and boarding risk C6, in addition to confidence level C1 and interest level C2. However, referring to interest level C2, originality C3, purpose of boarding C4, physical and mental state C5, and boarding risk C6 is not mandatory.

[0175] Here, the action suggestion unit 112A controls the suggestion of crew behavior through at least one of the content of the suggestion made by the action suggestion unit 112A (suggestion content J1), the method of presenting the suggestion J2, and motivational information J3 that encourages the adoption of the suggestion.

[0176] Control via proposed content J1 includes setting the content of the destination Lp and travel route Tp themselves, and setting the number of options (number of proposals) for the destination Lp and travel route Tp.

[0177] Control via presentation method J2 includes setting the amount of information when the proposed content J1 is visualized, setting the type of information, and setting the notification method for that information.

[0178] Here, setting the amount of information includes, for example, setting the number of characters when the proposed content J1 is written into text. Setting the type of information includes, for example, setting whether the proposed content J1 is displayed as text on the terminal 11's display or whether it is announced as audio from the terminal 11's speaker. Setting the notification method of the information includes setting whether or not it is necessary to highlight at least a part of the proposed content J1 that is displayed as text, and setting whether or not it is necessary to express it in a more friendly way to the passenger U. Expressing it in a friendly way can be done, for example, by writing it in a more colloquial way.

[0179] Control via motivational information J3 includes setting whether or not to present and the content of a first additional information corresponding to preference characteristics C7, setting whether or not to present and the content of a second additional information based on past riding history, and setting whether or not to present and the content of a third additional information based on the date and time of vehicle V use and weather conditions.

[0180] Here, the first additional information includes various tourist information such as places of interest, local specialties, and meals corresponding to preference trait C7. The second additional information includes information indicating past experiences of passenger U or other users U', such as memories of passenger U or other users, such as "Many users have been seen smiling at this destination in the past." The third additional information includes information indicating the relationship between the date and time of use or weather and the suggested destination Lp or travel route Tp, such as "We suggested a destination where you can see the night view because it's an anniversary" or "We suggested a coastal travel route because it's a sunny day."

[0181] (3-2. Processing related to confidence level) As illustrated in Figures 9 and 10, the action suggestion unit 112A controls the suggested content J1 and presentation method J2 according to the confidence level C1.

[0182] Specifically, as part of controlling the proposed content J1, the action suggestion unit 112A sets more crew action options to present to the passenger U when the confidence level C1 is low compared to when the confidence level C1 is high. More specifically, the action suggestion unit 112A sets more options for destination Lp and travel route Tp as the confidence level C1 decreases.

[0183] Furthermore, as part of controlling the proposed content J1, when the confidence level C1 is low, the action suggestion unit 112A suggests to the passenger U crew behaviors that are geared towards the general public, including other user groups U', compared to when the confidence level C1 is high.

[0184] This can be achieved by reducing the weighting (importance) of preference characteristics C7 when selecting travel activities from the destination database 301. Reducing the weighting can be rephrased as downplaying preference characteristics C7. Downplaying preference characteristics C7 can be rephrased as presenting a wider variety of options.

[0185] Furthermore, as part of controlling the presentation method J2, the action suggestion unit 112A executes notifications via the terminal 11 more concisely when the confidence level C1 is low compared to when the confidence level C1 is high. Specifically, the action suggestion unit 112A sets the amount of information notified via the terminal 11 to be smaller as the confidence level C1 decreases.

[0186] Furthermore, as part of the control of the presentation method J2, the action suggestion unit 112A executes notifications via the terminal 11 in a more user-friendly manner when the confidence level C1 is high, compared to when the confidence level C1 is low.

[0187] (3-3. Processing related to the level of interest) As illustrated in Figures 9 and 10, the behavior suggestion unit 112A controls the suggested content J1 and motivational information J3 according to the level of interest C2. The behavior suggestion unit 112A controls the suggestion of crew behavior according to the stage of behavioral change.

[0188] Specifically, as part of controlling the proposed content J1, the action proposal unit 112A differentiates whether or not to propose crew actions according to the level of interest C2. More specifically, the action proposal unit 112A proposes movement actions as crew actions only during the "action phase" and "maintenance phase" of the four stages: "precontemplation phase," "contemplation phase," "action phase," and "maintenance phase."

[0189] Here, the behavior suggestion unit 112A controls the suggested content J1 to be executed during the "execution phase" and the "maintenance phase" by reflecting the passenger U's preference characteristics C7 in the suggested passenger behavior according to the level of interest C2. Specifically, when the level of interest C2 is classified as the "maintenance phase," the behavior suggestion unit 112A proposes passenger U more general passenger behavior, including other user groups U', compared to when the level of interest C2 is classified as the "execution phase."

[0190] This can be achieved by reducing the weighting (importance) of preference characteristics C7 when selecting travel activities from the destination database 301. Reducing the weighting can be rephrased as downplaying preference characteristics C7. Downplaying preference characteristics C7 can be rephrased as presenting a wider variety of options.

[0191] Furthermore, as a process common to both the action suggestion unit 112A and the motion suggestion unit 112B described later, the suggestion unit 112, during the "precontemplation stage," periodically introduces the functions of the suggestion system 1001, such as presenting new functions of the suggestion system 1001, as part of the control of motivational information J3, but does not make suggestions for occupant actions or vehicle actions. The action suggestion unit 112A also, during the "precontemplation stage," displays past destinations Lp linked to the emotions of the passenger U as part of the control of motivational information J3 (a function to display the memories of the passenger U).

[0192] Furthermore, during the "contemplation stage," the proposal unit 112 only estimates individual characteristics C0 and does not propose crew behavior or vehicle movement. Specifically, during the "contemplation stage," the behavior proposal unit 112A performs at least one of the following: receiving responses to the second questionnaire A2 and analyzing individual characteristics C0 based on crew images. The same applies to the movement proposal unit 112B, which will be described later.

[0193] In addition, during the "maintenance phase," the proposal unit 112 performs gamification via the terminal 11 as a control of the motivational information J3. This gamification can be performed by rewarding the passenger U each time they accept a proposal from the terminal 11. This reward is represented as an internal parameter of the proposal system 1001. Depending on the reward, the system may also perform a process to change the screen within the terminal 11 (for example, to make the screen more elaborate).

[0194] (3-4. Processing related to uniqueness) As illustrated in Figures 9 and 10, the action suggestion unit 112A controls the suggested content J1 according to the uniqueness C3 of the passenger U.

[0195] Specifically, as part of controlling the proposed content J1, the action suggestion unit 112A reflects the passenger's preference characteristics C7 more significantly in the suggested crew actions when the originality C3 is high compared to when the originality C3 is low.

[0196] This can be achieved by increasing the weighting (importance) of preference characteristics C7 when selecting travel activities from the destination database 301. Increasing the weighting can be rephrased as giving more importance to preference characteristics C7. Giving more importance to preference characteristics C7 can be rephrased as presenting options tailored to the individual passenger U.

[0197] (3-5. Procedures related to purpose of boarding and physical and mental condition) As illustrated in Figures 9 and 10, the action suggestion unit 112A controls the suggested content J1, presentation method J2, and motivational information J3 by referring to the purpose of boarding C4 and the mental / physical state C5. Here, the action suggestion unit 112A controls the suggestion of crew behavior according to the category to which the purpose of boarding C4 belongs, such as sightseeing. The action suggestion unit 112A controls the suggestion of crew behavior according to the category to which the emotion in the mental / physical state C5 belongs, such as "joy (happiness)."

[0198] Specifically, the action suggestion unit 112A, as part of controlling the suggested content J1, makes suggestions for destination Lp according to the purpose of boarding C4 and the mental and physical state C5. The action suggestion unit 112A makes different suggestions for destination Lp depending on the purpose of boarding C4, such as "If the target area is ●● and the purpose of boarding is ▲▲, then ■■ will be suggested as the destination." This suggestion is effective when a specific area is answered as the destination Lp in the second questionnaire A2, or when the destination Lp is not answered.

[0199] Furthermore, as part of controlling the motivational information J3, the action suggestion unit 112A determines the destination Lp for the trip if the passenger U is traveling for sightseeing purposes, for example, based on the response to the second questionnaire A2. The action suggestion unit 112A then presents sightseeing information, such as landmarks around the destination Lp, to the passenger U via the terminal 11. This presentation may be done through photographs or videos.

[0200] Furthermore, as part of controlling the motivational information J3, the behavior suggestion unit 112A determines the destination Lp if the passenger U is traveling for purposes other than sightseeing, for example, based on the response to the second questionnaire A2. The behavior suggestion unit 112A presents the passenger U, via terminal 11, with information about the region to which the destination Lp belongs (regional information), such as scenic spots if the purpose is for funerals, or restaurants if the purpose is for business. This presentation may be done through photos or videos. The presentation of regional information can be performed based on the contents stored in the external server 300.

[0201] Furthermore, the action suggestion unit 112A varies the suggested destination Lp according to the passenger's mental and physical state C5. This allows for the suggestion of a destination Lp that is appropriate to the passenger U's emotions or physical condition.

[0202] Furthermore, as part of controlling the proposed content J1, the action suggestion unit 112A proposes a travel route Tp according to the purpose of boarding C4 and the mental and physical state C5. The action suggestion unit 112A varies the proposed travel route Tp according to the purpose of boarding C4, for example, "If the destination is ●● and the purpose of boarding is ▲▲, propose ■■ as the travel route." This proposal is effective when the travel route Tp is not answered in the second questionnaire A2.

[0203] Furthermore, the action suggestion unit 112A varies the suggested travel route Tp according to the mental and physical state C5. This allows for the suggestion of a travel route Tp that is appropriate to the passenger U's emotions or physical condition.

[0204] Furthermore, when proposing a travel route Tp, the action proposal unit 112A acquires fuel or power refueling points Li along the travel route Tp based on the travel information I3. These refueling points Li can be acquired, for example, using location information such as GPS. The action proposal unit 112A proposes stopping at these refueling points Li as crew actions (see the black circles in Figure 3B).

[0205] This suggestion is particularly effective, for example, when vehicle V is a rental car or a shared car. The action suggestion unit 112A determines whether vehicle V is a rental car or a shared car according to the usage flag, and if the determination is YES, it turns on the refueling point suggestion function. Alternatively, the action suggestion unit 112A may determine whether the purpose of boarding C4 is sightseeing, and if the determination is YES, it may turn on the refueling point suggestion function.

[0206] Furthermore, as part of controlling the presentation method J2, the action suggestion unit 112A changes at least one of the display color of the screen display and the voice quality of the voice according to the purpose of boarding C4 and the mental and physical state C5.

[0207] Here, the action suggestion unit 112A sets the voice quality to a subdued tone when the purpose of boarding C4 is classified as a funeral or the mental / physical state C5 is classified as sadness. On the other hand, the action suggestion unit 112A sets the voice quality to a tone with prominent intonation when the purpose of boarding C4 is classified as a celebration or the mental / physical state C5 is classified as joy.

[0208] Furthermore, when the purpose of boarding C4 is classified as a funeral or the mental / physical state C5 is classified as sadness, the action suggestion unit 112A sets the display color of the screen (for example, the display color of the text indicating the suggestion) to a more monochrome color. On the other hand, when the purpose of boarding C4 is classified as a celebration or the mental / physical state C5 is classified as joy, the action suggestion unit 112A sets the display color of the screen to a more colorful color.

[0209] (3-6. Handling of boarding risks) As illustrated in Figure 9, the action suggestion unit 112A controls at least one of the suggested content J1, presentation method J2, and motivational information J3 by referring to the boarding risk C6.

[0210] More specifically, the action suggestion unit 112A according to this embodiment changes the choice of destination Lp or route Tp based on the economic risk and safety risk that constitute the boarding risk C6. More specifically, the action suggestion unit 112A proposes a destination Lp or route Tp that is cheaper and has a lower safety risk. If there is a trade-off between economic risk and safety risk, the action suggestion unit 112A makes a proposal that prioritizes safety risk.

[0211] <4. Terminal Control Unit> The terminal control unit 113 notifies the terminal 11 of the proposal decided by the action proposal unit 112A. Depending on the passenger U's response to the notification, the proposal from the action proposal unit 112A is accepted or rejected, and factors influencing the proposal from the action proposal unit 112A, such as confidence level C1, are updated accordingly.

[0212] <5. Motion proposal section> The operation suggestion unit 112B according to this embodiment is composed of a machine learning model. The machine learning model that constitutes the operation suggestion unit 112B may be a neural network or a rule-based model. It is not essential that the operation suggestion unit 112B be composed of a machine learning model.

[0213] (5-1.Basic concept) The motion suggestion unit 112B proposes the aforementioned defined mobile vehicle operations to the passenger U. In this embodiment, the motion suggestion unit 112B proposes to the passenger U, as mobile vehicle operations, sound control of the vehicle V, lighting control of the vehicle V, or air conditioning control of the vehicle V.

[0214] As an example, the operation suggestion unit 112B performs the following as sound control of the vehicle V: setting up the sound equipment 123 and operating the sound equipment 123. More specifically, the setting of the sound equipment 123 includes one or more of the genre of music to be played in the vehicle, the title of the music, and the volume of the music. More specifically, the setting of the sound equipment 123 further includes determining whether or not to play a playlist, the titles of the songs that make up the playlist, and setting whether or not to enable noise cancellation.

[0215] More specifically, as illustrated in Figure 9, the motion suggestion unit 112B determines the proposed mobile operation (e.g., acoustic control) for the passenger U based on the preference characteristics C7 estimated from the responses to the second questionnaire A2. More specifically, when determining the acoustic control proposed for the passenger U, the motion suggestion unit 112B refers to the aforementioned acoustic database 302 stored in the external server 300. This acoustic database 302 contains a database of acoustic controls corresponding to the preference characteristics C7.

[0216] The mobile body operation determined by the operation proposal unit 112B is proposed to the passenger U via the terminal 11. At that time, the operation proposal unit 112B according to this embodiment controls the proposal of the mobile body operation based on the confidence level C1 and interest level C2 in the personal characteristics C0.

[0217] In detail, the motion suggestion unit 112B controls the suggestion of mobile body movements by referring to reliability C1 and interest level C2, as well as originality C3, purpose of boarding C4, physical and mental state C5, and boarding risk C6. However, referring to interest level C2, originality C3, purpose of boarding C4, physical and mental state C5, and boarding risk C6 is not mandatory.

[0218] Here, the action suggestion unit 112B controls the suggestion of crew behavior through at least one of the following: the content of the suggestion made by the action suggestion unit 112B (suggestion content J1), the method of presenting the suggestion J2, and motivational information J3 that encourages the adoption of the suggestion.

[0219] The control via proposed content J1 includes setting the content of sound control, such as song selection and volume, and setting the number of options (number of proposals) for the proposed content.

[0220] Control via presentation method J2 includes setting the amount of information when the proposed content J1 is visualized, setting the type of information, and setting the notification method for that information.

[0221] Here, setting the amount of information includes, for example, setting the number of characters when the proposed content J1 is written into text. Setting the type of information includes, for example, setting whether the proposed content J1 is displayed as text on the terminal 11's display or whether it is announced as audio from the terminal 11's speaker. Setting the notification method of the information includes setting whether or not it is necessary to highlight at least a part of the proposed content J1 that is displayed as text, and setting whether or not it is necessary to express it in a more friendly way to the passenger U. Expressing it in a friendly way can be done, for example, by writing it in a more colloquial way.

[0222] Control via motivational information J3 will be discussed later in relation to the level of interest C2.

[0223] (5-2. Processing related to confidence level) As illustrated in Figures 9 and 10, the operation suggestion unit 112B controls the suggested content J1 and the presentation method J2 according to the confidence level C1.

[0224] Specifically, as part of controlling the proposed content J1, the motion suggestion unit 112B sets more options for mobile body movements to be presented to the passenger U when the reliability C1 is low compared to when the reliability C1 is high. For example, when the reliability C1 is high, the motion suggestion unit 112B reduces the number of songs suggested to the passenger U compared to when the reliability C1 is low. More specifically, the motion suggestion unit 112B sets more options for sound control as the reliability C1 decreases.

[0225] Furthermore, as part of controlling the proposed content J1, the operation proposal unit 112B determines whether or not to propose playback of a playlist containing multiple songs, and adjusts the number of songs that make up the playlist, based on the reliability level C1. More specifically, the operation proposal unit 112B proposes playback of the playlist when the reliability level C1 exceeds a predetermined second threshold. More specifically, the operation proposal unit 112B sets a larger number of songs to make up the playlist as the reliability level C1 increases.

[0226] Furthermore, as part of the control of the proposed content J1, the operation proposal unit 112B reflects the preference characteristics C7 more significantly in the proposed movement of the mobile body when the reliability C1 is high compared to when the reliability C1 is low.

[0227] This can be achieved by increasing the weighting (importance) of preference characteristic C7 when selecting acoustic controls from the acoustic database 302. Increasing the weighting can be rephrased as giving more importance to preference characteristic C7. Giving more importance to preference characteristic C7 can be rephrased as presenting options tailored to the individual passenger U.

[0228] Furthermore, as part of the control of the presentation method J2, the operation suggestion unit 112B executes notifications via the terminal 11 more concisely when the confidence level C1 is low compared to when the confidence level C1 is high. Specifically, the operation suggestion unit 112B sets the amount of information notified via the terminal 11 to be smaller as the confidence level C1 decreases.

[0229] Furthermore, as part of the control of the presentation method J2, the operation suggestion unit 112B performs notifications via the terminal 11 in a more user-friendly manner when the confidence level C1 is high, compared to when the confidence level C1 is low.

[0230] (5-3. Processing related to the level of interest) As illustrated in Figures 9 and 10, the motion suggestion unit 112B controls the suggested content J1 and motivational information J3 according to the level of interest C2. The motion suggestion unit 112B controls the suggestion of mobile body movements according to the stage of behavioral change.

[0231] Specifically, as part of controlling the proposed content J1, the motion proposal unit 112B determines whether or not to propose mobile body movements depending on the level of interest C2. More specifically, the motion proposal unit 112B executes proposals for acoustic control as mobile body movements only during the "execution phase" and "maintenance phase," out of the four phases: "precontemplation phase," "contemplation phase," "execution phase," and "maintenance phase."

[0232] Here, the motion suggestion unit 112B, as a control of the suggested content J1 executed during the "execution phase" and the "maintenance phase," reflects the passenger U's preference characteristics C7 in the suggested mobile operation according to the level of interest C2. Specifically, when the level of interest C2 is classified as the "maintenance phase," the motion suggestion unit 112B proposes mobile operation to the passenger U that is geared towards the general public, including other user groups U', compared to when the level of interest C2 is classified as the "execution phase."

[0233] This can be achieved by reducing the weighting (importance) of preference characteristic C7 when selecting acoustic controls from the acoustic database 302. Reducing the weighting can be rephrased as downplaying preference characteristic C7. Downplaying preference characteristic C7 can be rephrased as presenting a wider variety of options.

[0234] (5-4. Processing related to uniqueness) As illustrated in Figures 9 and 10, the operation suggestion unit 112B controls the suggested content J1 according to the uniqueness C3 of the passenger U.

[0235] Specifically, as part of the control of the proposed content J1, the motion suggestion unit 112B reflects the passenger's preference characteristics C7 more significantly in the proposed motion of the mobile body when the originality C3 is high compared to when the originality C3 is low.

[0236] This can be achieved by increasing the weighting (importance) of preference characteristic C7 when selecting acoustic controls from the acoustic database 302. Increasing the weighting can be rephrased as giving more importance to preference characteristic C7. Giving more importance to preference characteristic C7 can be rephrased as presenting options tailored to the individual passenger U.

[0237] (5-5. Processing related to purpose of boarding and physical and mental condition) As illustrated in Figures 9 and 10, the motion suggestion unit 112B controls the suggested content J1, presentation method J2, and motivational information J3 by referring to the purpose of boarding C4 and the mental / physical state C5. Here, the motion suggestion unit 112B controls the suggestion of vehicle movements according to the category to which the purpose of boarding C4 belongs, such as sightseeing. The motion suggestion unit 112B also controls the suggestion of vehicle movements according to the category to which the emotion in the mental / physical state C5 belongs, such as "joy (happiness)."

[0238] Specifically, the operation suggestion unit 112B executes an audio control suggestion according to the purpose of boarding C4 and the mental / physical state C5 as part of the control of the suggested content J1. The operation suggestion unit 112B changes one or more of the following according to the purpose of boarding C4 or the mental / physical state C5: the genre of music to be played in the vehicle, the song titles of the music, and the volume of the music.

[0239] For example, if the purpose of boarding C4 is sightseeing or a celebratory occasion, or if the mental and physical state C5 is "joy (happiness)" or "fun", the action suggestion unit 112B will suggest a more upbeat sound, a song belonging to the pop genre, or a louder volume.

[0240] On the other hand, if the purpose of boarding C4 is business or funeral, or if the mental / physical state C5 is "sadness" or "anger," the action suggestion unit 112B will suggest slower-tempo sounds, songs belonging to classical or folk music, or lower volume levels.

[0241] In particular, when the purpose of boarding (C4) is business, if multiple passengers (U) are identified, a lower volume should be suggested to avoid disrupting conversations between the passengers (U).

[0242] Furthermore, as part of the control of the presentation method J2, the action suggestion unit 112B changes at least one of the display color of the screen display and the voice quality of the voice according to the purpose of boarding C4 and the mental and physical state C5.

[0243] Here, the action suggestion unit 112B sets the voice quality to a voice with subdued intonation when the purpose of boarding C4 is classified as a funeral or the mental / physical state C5 is classified as sadness. On the other hand, the action suggestion unit 112B sets the voice quality to a voice with prominent intonation when the purpose of boarding C4 is classified as a celebration or the mental / physical state C5 is classified as joy.

[0244] Furthermore, when the purpose of boarding C4 is classified as a funeral or the mental / physical state C5 is classified as sadness, the action suggestion unit 112B sets the display color of the screen (for example, the display color of the text indicating the suggestion) to a more monochrome color. On the other hand, when the purpose of boarding C4 is classified as a celebration or the mental / physical state C5 is classified as joy, the action suggestion unit 112B sets the display color of the screen to a more colorful color.

[0245] (5-6. Handling of boarding risks) As illustrated in Figure 9, the action suggestion unit 112B controls at least one of the suggested content J1, presentation method J2, and motivational information J3 by referring to the boarding risk C6.

[0246] More specifically, the operation suggestion unit 112B according to this embodiment changes the sound control options based on the safety risks that constitute the boarding risk C6. More specifically, when the safety risk is high, the operation suggestion unit 112B changes the content of the suggestion to the passenger U so that the volume is suppressed compared to when the safety risk is low.

[0247] <6. Operation Control Unit> The motion control unit 114 notifies the terminal 11 of the proposal decided by the motion proposal unit 112B (see the arrow in Figure 3A). Depending on the passenger U's response to the notification, the proposal from the motion proposal unit 112B is accepted or rejected. The motion proposal unit 112B causes the onboard computer 100 and then the vehicle V to execute the accepted motion via the actuator. In addition, depending on whether the proposal is accepted or rejected, factors that influence the proposal from the motion proposal unit 112B, such as the reliability C1, are updated accordingly.

[0248] [C. Preferential Treatment Determination Device] Figure 11 is a block diagram illustrating an example configuration of the preferential treatment determination device 1002. Figure 12 is a functional block diagram illustrating the configuration of the preferential treatment determination device 1002. Figure 13A is a graph illustrating the relationship between reliability and the first indicator K1. Figure 13B is a graph illustrating the relationship between boarding history and the first indicator K1. Figure 13C is a graph illustrating the relationship between relationship information L3 and the first indicator K1. Figure 14 is a graph illustrating the relationship between influence information L4 and the second indicator K2. Figure 15 is a graph illustrating the relationship between the first indicator K1, the second indicator K2, or sales promotion level N2 and preferential treatment level N1.

[0249] <1.Device configuration> As illustrated in Figure 11, the preferential treatment determination device 1002 is a computer in which the control unit 501 and the input / output interface (I / O interface) 502 are interconnected via a system bus.

[0250] The control unit 501 includes a processor 501a that executes various programs, a read-only ROM 501b that stores the BIOS and the like, and RAM 501c that functions as main memory.

[0251] The input / output interface 502 is connected to the operation unit 511, the display unit 512, the storage unit 513, and the communication unit 514.

[0252] The control unit 511 accepts operations and input from the operator. The control unit 511 is composed of at least one of a keyboard and a pointing device. The pointing device is, for example, a mouse.

[0253] The display unit 512 displays various information to the user, such as the results of program execution by the control unit 501. The display unit 512 is composed of a display such as a liquid crystal display or an organic EL display.

[0254] Storage 513 stores various information related to the preferential treatment level N1, which will be described later, and various programs for processing said information. Storage 513 is composed of SSDs (Solid State Drives), HDDs (Hard Disk Drives), etc.

[0255] The communication unit 514 is connected to the proposed system 1001 of each of the multiple vehicles V based on a predetermined communication standard. The communication unit 514 and each proposed system 1001 are connected in a way that enables data transmission and reception. Through this connection, the communication unit 514 obtains reliability data L1 and boarding history data L2 from the onboard computer 100 of each vehicle V.

[0256] Here, confidence data L1 is electronic data representing the confidence level C1 estimated by the estimation unit 111. Confidence data L1 is acquired separately for each passenger U in each vehicle V.

[0257] Furthermore, the boarding history data L2 is electronic data showing the boarding history of passenger U. Boarding history data L2 is acquired separately for each passenger U in each vehicle V. Boarding history data L2 can be constructed based on an index that represents the relationship between passenger U and vehicle V, which was referenced when estimating preference characteristics C7.

[0258] More specifically, the boarding history data L2 according to this embodiment consists of at least one of the following: the boarding frequency of passenger U, the cumulative boarding time of passenger U, and the cumulative distance traveled by passenger U in vehicle V. This information can be obtained based on electrical signals (detection signals) input from the movement information detection device 14 of each vehicle V.

[0259] The communication unit 514 is also connected to the internet Ne2, as previously mentioned. Through this connection, the communication unit 514 obtains relationship information L3 that characterizes the depth of the relationship between the passenger U and the manufacturing brand. Relationship information L3 is obtained separately for each passenger U in each vehicle V.

[0260] More specifically, the relationship information L3 according to this embodiment consists of at least one of the following: information on passenger U's visit to a dealership of vehicle V; information on visits to a website related to the manufacturing brand; participation history in official events related to the manufacturing brand; and SNS (Social Networking Service) information related to the manufacturing brand.

[0261] More specifically, the SNS information consists of at least one of the following: information indicating whether or not the user follows the official accounts of Vehicle V, the manufacturing brand, or the manufacturer; and the frequency or number of positive posts related to Vehicle V, the manufacturing brand, or the manufacturer. "Positive posts" here include posts supporting Vehicle V, the manufacturing brand, or the manufacturer.

[0262] Below, "rich social media information" is defined as following official accounts, having a high frequency of positive posts, or a large number of positive posts. Similarly, "poor social media information" is defined as not following official accounts, having a low frequency of positive posts, or a small number of positive posts.

[0263] Furthermore, the aforementioned store visit information includes the visit history of passenger U to the aforementioned store.

[0264] Here, the term "visit history" is used in a broad sense. "Visit history" here consists of at least one of the following: the frequency of visits by passenger U and the number of visits by passenger U. Hereafter, a high frequency of visits or a large number of visits will be defined as "a high visit history," and a low frequency of visits or a small number of visits will be defined as "a low visit history."

[0265] In addition, the store visit information may further include the response content to the questionnaire conducted at the store and supplementary information (information provided by the salesperson of the store regarding each passenger U) provided by the salesperson of the store for each passenger U. For example, when calculating the value of the first index K1 described later, the value of the first index K1 calculated by referring to the store visit history may be adjusted based on the response content and the supplementary information.

[0266] On the other hand, the visit information includes the visit history to the website.

[0267] Here, the term "visit history" is used in a broad sense, similar to the term "store visit history". The "visit history" referred to here is composed of at least one of the visit frequency of passenger U and the number of visits of passenger U. Hereinafter, a high visit frequency or a large number of visits is defined as "a large visit history", and a low visit frequency or a small number of visits is defined as "a small visit history".

[0268] In addition, the visit information may further include the transition history between pages on the website, the viewing time of each page, and the response content to the questionnaire conducted on the website. For example, the more transition history there is or the longer the viewing time is, the more it is regarded as "a large visit history", and the less transition history there is or the shorter the viewing time is, the more it is regarded as "a small visit history". For example, when calculating the value of the first index K1 described later, the value of the first index K1 calculated by referring to the visit history may be adjusted based on the response content.

[0269] Similarly, the term "participation history" is used in a broad sense. The "participation history" referred to here is composed of at least one of the participation frequency of passenger U and the number of participations of passenger U. Hereinafter, a high participation frequency or a large number of participations is defined as "a large participation history", and a low participation frequency or a small number of participations is defined as "a small participation history".

[0270] The communications unit 514 also obtains influence information L4, which characterizes the social influence of passenger U, via a connection to the internet Ne2. Influence information L4 is obtained separately for each passenger U in each vehicle V.

[0271] More specifically, the influence information L4 according to this embodiment consists of at least one of the following: a history of exposure in mass media and activity status on social media.

[0272] The term "exposure history" is used in a broad sense. Herein, "exposure history" consists of at least one of the following: the frequency of exposure of passenger U and the number of times passenger U has been exposed. Hereafter, a high frequency of exposure or a large number of exposures will be defined as "a high exposure history," and a low frequency of exposure or a small number of exposures will be defined as "a low exposure history."

[0273] More specifically, social media activity includes at least one of the following: the posting history of passenger U, which was used as the target of the first indicator K1 described below; the number of followers of said passenger U; the number of impressions of said passenger U; and whether or not monetization is possible on social media.

[0274] Furthermore, the "posting history" that constitutes influence information L4 differs from the SNS information that constitutes relationship information L3, and includes all posts other than those related to vehicle V, manufacturing brand, or manufacturer.

[0275] In the following, "rich social media activity information" is defined as a high frequency of posts, a high number of posts, a large number of followers or impressions, or permission to monetize social media. Similarly, "poor social media activity information" is defined as a low frequency of posts, a low number of posts, a small number of followers or impressions, or permission to monetize social media.

[0276] As illustrated in Figure 12, the preferential treatment determination device 1002 according to this embodiment includes a reliability acquisition unit 521, a boarding history acquisition unit 522, a relationship acquisition unit 523, a first calculation unit 524, an influence acquisition unit 531, a second calculation unit 532, a preferential treatment determination unit 541, a candidate selection unit 542, a benefit setting unit 543, and a token granting unit 544. These elements are configured, for example, by the control unit 501 of the preferential treatment determination device 1002.

[0277] <2. Calculation of the first indicator> As shown in Figure 12, the reliability acquisition unit 521 acquires reliability data L1 indicating the reliability C1 of passenger U via the communication unit 514. The boarding history acquisition unit 522 acquires boarding history data L2 indicating the boarding history of the same passenger U via the communication unit 514. The relationship acquisition unit 523 acquires relationship information L3 for the same passenger U via the communication unit 514.

[0278] As shown in Figure 12, the first calculation unit 524 calculates the value of the first index K1 based on the confidence level C1 estimated by the estimation unit 111, specifically the confidence level data L1 acquired by the confidence level acquisition unit 521.

[0279] The first indicator K1 can be called an engagement indicator that shows the strength of the connection between each passenger U and the vehicle V, the manufacturing brand, or the manufacturer. The calculation of the first indicator K1 can be performed, for example, based on a first map 591 which defines the relationship between various variables F, including confidence level C1, and the value of the first indicator K1. This first map 591 is stored, for example, in storage 513. A machine learning model may be used instead of the first map 591.

[0280] Here, the value of the first index K1 is specified to increase or decrease according to the confidence level C1. Specifically, as illustrated in graph G11 of Figure 13A, the value of the first index K1 is specified to increase as the confidence level C1 increases (the more confident the result).

[0281] Furthermore, the first calculation unit 524 calculates the value of the first index K1 by referring to the confidence level C1 and the boarding history of the same passenger U (specifically, boarding history data L2) as the target of the estimation of the confidence level C1, so as to increase or decrease according to the confidence level C1 and the boarding history.

[0282] Here, the value of the first indicator K1 is specified to increase or decrease according to the passenger's flight history. More specifically, the value of the first indicator K1 is specified to increase as the passenger U's flight history becomes more extensive.

[0283] More specifically, the value of the first indicator K1 is calculated by referring to at least one of the following: the passenger U's boarding frequency, the passenger U's cumulative boarding time, and the cumulative distance traveled by passenger U on vehicle V.

[0284] Specifically, the value of the first indicator K1 is set to increase as the frequency of passenger U's travel increases, or as the cumulative value of travel time or distance traveled increases, as illustrated in graph G12 of Figure 13B.

[0285] Furthermore, the first calculation unit 524 calculates the value of the first index K1 by referring to the confidence level C1, the boarding history data L2, and the relationship information L3 of the same passenger U that is the target of the estimation for the confidence level C1, so that it increases or decreases according to the confidence level C1, the boarding history data L2, and the relationship information L3. Referring to the boarding history data L2 is not mandatory.

[0286] Here, the value of the first indicator K1 is specified to increase or decrease according to the relationship information L3. More specifically, the value of the first indicator K1 is specified to increase as the relationship between the passenger U and the manufacturing brand deepens.

[0287] More specifically, the value of the first indicator K1 is calculated by referring to at least one of the following: information on passenger U's visits to the vehicle V dealership, information on visits to the manufacturer brand's website, participation history in official events related to the manufacturer brand, and social media information related to the manufacturer brand.

[0288] Specifically, as illustrated in the graph G13 of FIG. 13C, the value of the first index K1 is defined to increase as the store visit history of the passenger U is more frequent, the visit history is more frequent, the participation history is more frequent, or the SNS information is more abundant.

[0289] <3. Calculation of the Second Index> As shown in FIG. 12, the influence acquisition unit 531 acquires the influence information L4 of the same passenger U as the estimation target of the reliability C1 of the passenger U via the communication unit 514.

[0290] As shown in FIG. 12, the second calculation unit 532 calculates the value of the second index K2 based on the influence information L4 acquired by the influence acquisition unit 531.

[0291] The second index K2 can be called an influence index indicating the strength of the social influence of each passenger U. The calculation of the second index K2 can be executed based on, for example, a second map 592 in which the relationship between the variables related to the influence information L4 and the value of the second index K2 is defined. This second map 592 is stored, for example, in the storage 513. Instead of the second map 592, a machine learning model may be used.

[0292] Here, the value of the second index K2 is defined to increase or decrease according to the strength of the social influence as described above. Specifically, the value of the second index K2 is defined to increase as the social influence is stronger.

[0293] Specifically, as illustrated in the graph G21 of FIG. 14, the value of the second index K2 is defined to increase as the exposure history to the mass media is more frequent or the activity status on the social media is more abundant.

[0294] <4. Scoring of the Preferential Treatment> As shown in FIG. 12, the preferential treatment determination unit 541 scores the preferential treatment N1 of the passenger U for which the first index K1 is the estimation target among the plurality of passengers U based on the value of the first index K1 calculated by the first calculation unit 524.

[0295] Here, the value of preferential treatment N1 is defined to increase or decrease according to the first indicator K1. The value of preferential treatment N1 is associated with each of multiple passengers U and indicates the priority level of each passenger U. Passenger U with a high value of preferential treatment N1 can be considered a customer that deserves higher priority than passenger U with a low value of preferential treatment N1.

[0296] More specifically, the value of the preferential treatment score N1 is defined to be broadly monotonically increasing with respect to the first indicator K1, as illustrated in graph G31 of Figure 15. In other words, the value of the preferential treatment score N1 is defined to increase as the value of the first indicator K1 increases, or to remain constant even as the value of the first indicator K1 increases.

[0297] More specifically, the value of the preferential treatment score N1 is specified to increase or decrease in accordance with both the first indicator K1 and the second indicator K2. The value of the preferential treatment score N1 is specified to increase broadly monotonically with respect to the second indicator K2, as illustrated in graph G31 in Figure 15. In other words, the value of the preferential treatment score N1 is specified to increase as the value of the second indicator K2 increases, or to maintain a constant value even if the value of the second indicator K2 increases.

[0298] More specifically, the value of the preferential treatment level N1 according to this embodiment is defined to increase or decrease in accordance with the sales promotion level N2, which is based on both the first indicator K1 and the second indicator K2. The preferential treatment level determination unit 541 according to this embodiment uses the product of the first indicator K1 and the second indicator K2 for the sales promotion level N2. The value of the preferential treatment level N1 is defined to increase broadly monotonically with respect to the sales promotion level N2, as illustrated in graph G31 of Figure 15. That is, the value of the preferential treatment level N1 is defined to increase as the value of the sales promotion level N2 increases, or to maintain a constant value even if the value of the sales promotion level N2 increases. In particular, in this embodiment, the value of the preferential treatment level N1 is defined to plateau when the value of the sales promotion level N2 exceeds a positive predetermined threshold Th.

[0299] Specifically, the calculation of the preferential treatment level N1 in this embodiment can be performed, for example, based on a third map 593 that defines the relationship between the value of the sales promotion level N2 and the value of the preferential treatment level N1. This third map 593 is stored, for example, in storage 513. A machine learning model may be used instead of the third map 593.

[0300] Alternatively, instead of using the product of the first indicator K1 and the second indicator K2 for the sales promotion score N2, the sum of the first indicator K1 and the second indicator K2 may be used. The sales promotion score N2 only needs to be defined as broadly monotonically increasing with respect to both the first indicator K1 and the second indicator K2.

[0301] <5. Specific examples of the procedure for determining the degree of preferential treatment> Figure 16 is a flowchart illustrating the procedure for determining the degree of preferential treatment. Steps S502 and S503 may be executed simultaneously, or their execution order may be reversed.

[0302] First, in step S501 of Figure 16, the reliability acquisition unit 521 acquires reliability data L1, the boarding history acquisition unit 522 acquires boarding history data L2, the relationship acquisition unit 523 acquires relationship information L3, and the influence acquisition unit 531 acquires influence information L4.

[0303] In the following step S502, the first calculation unit 524 calculates the first index K1 based on the confidence data L1, the boarding history data L2, the relationship information L3, and the first map 591.

[0304] In the following step S503, the second calculation unit 532 calculates the second indicator K2 based on the influence information L4 and the second map 592.

[0305] In the following step S504, the preferential treatment determination unit 541 calculates the sales promotion degree N2 based on the first indicator K1 and the second indicator K2.

[0306] In the following step S505, the preferential treatment determination unit 541 determines the value of the preferential treatment N1 based on the sales promotion level N2 and the third map 593.

[0307] The preferential treatment determination device 1002 performs the processes from step S501 to step S505 for each of the multiple passengers U. The preferential treatment determination device 1002 assigns a preferential treatment value N1 to each passenger U.

[0308] Furthermore, each parameter contributing to the preferential treatment level N1, such as the reliability level C1, can change in real time based on the relationship between the passenger U and the vehicle V. The preferential treatment determination device 1002 periodically executes the processes from steps S501 to S505. As a result, the preferential treatment level N1 for each passenger U is also updated periodically.

[0309] <6. Processing based on preferential treatment> Figure 17 is a diagram illustrating the selection process for preferred customers, Us. Figure 18 is a diagram illustrating the digital token To that is awarded to preferred customers, Us.

[0310] As shown in Figure 12, the candidate selection unit 542 performs processing based on the preferential treatment level N1 value determined by the preferential treatment level determination unit 541.

[0311] Specifically, the candidate selection unit 542 according to this embodiment selects and outputs passengers who are eligible for preferential treatment from among multiple passengers U, based on the level of the preferential treatment score N1 scored by the preferential treatment determination unit 541.

[0312] As an example, the candidate selection unit 542 classifies passengers U whose preferential treatment value N1 is relatively high into preferred customers Us, as illustrated in Figure 17. The candidate selection unit 542 generates a preferred customer list N3, which lists the preferred customers Us, and outputs it (see Figure 12). The preferred customer list N3 is output as electronic data in CSV format, for example.

[0313] Furthermore, the determination of whether the value of preferential treatment N1 is ranked highly may be made based on whether the value of preferential treatment N1 is above a predetermined third threshold, or it may be made using statistical indicators based on each preferential treatment N1, such as the standard score of each preferential treatment N1.

[0314] The reward setting unit 543 then sets the rewards to be given to multiple passengers U according to the value of the preferential treatment level N1. For example, as shown in Figure 17, the reward setting unit 543 in this embodiment gives one or more rewards to passengers U classified as preferred customers Us. The reward setting unit 543 does not give any rewards to passengers U that are not classified as preferred customers Us.

[0315] Information indicating whether or not each passenger U is granted a prescribed benefit is output as a benefit grant list N4, which is linked to or integrated with the preferred customer list N3 (see Figure 12). The benefit grant list N4 is output as electronic data in CSV format, for example.

[0316] More specifically, as illustrated in Figure 17, the benefits setting unit 543 grants at least one of the following benefits to passengers U who are designated as eligible for preferential treatment, i.e., preferred customers Us: the benefit of lending a vehicle V; the benefit of participating in official events related to the manufacturing brand of vehicle V; the benefit of granting exclusive functions to vehicle V; and a monetary benefit when purchasing or selling vehicle V.

[0317] Limited features are, for example, limited features of SDV. The cost of granting limited features as a benefit can be discounted or waived depending on the preferential treatment level N1. Monetary benefits are, for example, new car discount coupons. In addition, the rental fee for rental benefits can be discounted or waived depending on the preferential treatment level N1, or the rental time can be extended depending on the preferential treatment level N1.

[0318] The preferential treatment determination device 1002 transmits the preferential customer list N3 and the benefit granting list N4 to the public relations department of the manufacturer, or to sales companies and dealers, via network communication such as the Internet Ne2.

[0319] The candidate selection unit 542 further certifies passengers U from among the preferred customers Us whose preferential treatment level N1 is above a predetermined fourth threshold (threshold) as influencers Ui. The preferential treatment determination device 1002 grants further benefits to influencers Ui via the benefit setting unit 543, or sends a list of influencers Ui to the public relations department of the manufacturer to propose advertising plans with those influencers Ui.

[0320] The token granting unit 544 grants the preferred customer Us a digital token To, which is managed on the blockchain, as illustrated in Figure 18. The digital token To is, for example, an NFT (Non-Fungible Token). The value of this digital token To fluctuates according to the values ​​of the first indicator K1, the second indicator K2, the preferential treatment level N1, or the sales promotion level N (see graph G41 in Figure 18).

[0321] The reward setting unit 543 then sets different rewards for preferred customers Us according to the value of the digital token To. The setting target according to the value of the digital token To may be the granting fee for the limited function reward, the discount amount for the new car discount coupon, the amount of the loan fee, or the length of the loan period. In addition, the setting of whether or not to grant each of the rewards exemplified in Figure 17 can be set individually according to the value of the digital token To.

[0322] <7. Specific examples of processing based on preferential treatment> Figure 19 is a flowchart illustrating a process based on the degree of preferential treatment.

[0323] First, in step S601 in Figure 19, the candidate selection unit 542 reads the value of preferential treatment N1 for a predetermined passenger U. In the following step S602, the candidate selection unit 542 determines whether the value of preferential treatment N1 read in step S601 belongs to the higher category. As mentioned above, this determination may be made based on a comparison between the value of preferential treatment N1 and the third threshold, or it may be made based on a statistical indicator related to preferential treatment N1.

[0324] If the determination in step S601 is NO, the candidate selection unit 542 does not select passenger U, who was selected in step S601 for reading the preferential treatment level N1, as a preferred customer Us, and terminates the process shown in Figure 19. On the other hand, if the determination in step S602 is YES, the candidate selection unit 542 proceeds to step S603 of the control process.

[0325] In step S603, the candidate selection unit 542 classifies (selects) the passenger U, who was selected in step S601 for preferential treatment level N1, into preferred customers Us.

[0326] In the following step S604, the token granting unit 544 grants the digital token To to passenger U, who is classified as a preferred customer Us.

[0327] In the subsequent step S605, the candidate selection unit 542 determines whether the value of the preferential treatment N1 read in step S601 is equal to or greater than a predetermined fourth threshold (threshold).

[0328] If the determination in step S605 is NO, the candidate selection unit 542 skips the following step S606 and proceeds to step S607 of the control process. On the other hand, if the determination in step S605 is YES, the candidate selection unit 542 proceeds to step S606 of the control process.

[0329] In step S606, the candidate selection unit 542 classifies (selects) the passenger U, who was selected in step S601 as a target for reading with preferential treatment level N1, as an influencer Ui.

[0330] In the following step S604, the reward setting unit 543 grants a reward to passenger U, who is classified as a preferred customer Us or influencer Ui, and sets the details of that reward.

[0331] <8. Other processes> In addition, the preferential treatment determination unit 541 may aggregate the preferential treatment level N1 for each region where each passenger U is located, and determine the level of preferential treatment level N1 for each region using a statistical indicator based on the aggregated results. In this case, the statistical indicator may be the sum of the preferential treatment levels N1 for each passenger U in each region, or the average value of the preferential treatment levels N1 for each passenger U in each region. By determining the level of preferential treatment level N1 for each region, it can be used to adjust the number of units shipped to each region.

[0332] [D. Effects, etc.] In recent years, there has been an increasing number of vehicles (V) equipped with functions that provide various services, such as suggesting destinations (Lp) according to the passenger's purpose of travel (C4), suggesting routes (Tp) to destinations (Lp), and suggesting music to be played in the vehicle.

[0333] On the other hand, among the general passengers U, including the driver of vehicle V, there may be passengers U who feel aversion to the services provided by vehicle V. Such aversion is reflected in the passengers U's responses to the services provided, such as the frequency with which they accept or reject the services. The inventors of this application hypothesize that this aversion fluctuates depending on the level of trust C1 between passenger U and vehicle V, the services, or the manufacturing brand. If a high level of trust has been established between passenger U and the manufacturing brand, it is thought that the aversion to the services provided will also be mitigated in proportion to that level of trust.

[0334] In other words, the inventors of this invention believe that each ordinary passenger U has their own level of trust C1 in the vehicle V, the service, or the manufacturing brand, and that this level of trust C1 is reflected in the passenger U's response to the services provided by the vehicle V.

[0335] Furthermore, the higher the confidence level C1, the stronger the "connection" between the passenger U, the vehicle V, the services offered, or the manufacturing brand. The strength of the connection can be quantified according to the level of confidence level C1. By giving preferential treatment to passenger U with a strong connection, it is possible to achieve more effective PR compared to passenger U with a weak connection.

[0336] Specifically, the preferential treatment determination system S according to this embodiment estimates a confidence level C1 based on the passenger U's response to the services provided, as illustrated in Figures 4 and 5, and calculates a value for the first index K1, which indicates the strength of the association, based on that confidence level C1.

[0337] The preferential treatment determination system S, as illustrated in Figures 12, 13A, 16, and 17, scores the preferential treatment level N1 of passenger U based on the value of the first indicator K1, and selects passenger U based on that preferential treatment level N1. This configuration allows for more appropriate selection of preferred customers Us.

[0338] Furthermore, as illustrated in Figures 12, 13B, and 16, the preferential treatment determination system S can reflect the "strength of the passenger's attachment" to the vehicle V when calculating the first indicator K1 by referring to the boarding history (boarding history data L2) in addition to the confidence level C1. This is advantageous in more appropriately selecting preferential customers Us.

[0339] Furthermore, as illustrated in Figure 13B, by referring to boarding frequency, boarding time, and mileage as part of the boarding history, the "strength of attachment" to vehicle V can be more appropriately reflected when calculating the value of the first indicator K1. This will be advantageous in more appropriately selecting preferred customers Us.

[0340] Furthermore, as illustrated in Figures 12, 13C, and 16, the preferential treatment determination system S, by referring to relationship information L3 in addition to confidence level C1, can reflect the "strength of the connection" between passenger U and the manufacturing brand from a different perspective than confidence level C1 when calculating the first indicator K1. This is advantageous in more appropriately selecting preferred customers Us.

[0341] Furthermore, as illustrated in Figure 13C, by referring to relationship information L3, such as store visit information, visit history, participation history, and SNS information, the "strength of ties" to the manufacturing brand can be more appropriately reflected when calculating the value of the first indicator K1. This is advantageous in more appropriately selecting preferred customers Us.

[0342] Furthermore, as illustrated in Figures 12, 14, and 16, the preferential treatment determination system S, in scoring the preferential treatment level N1, refers to a second indicator K2 that considers the strength of the passenger U's own social influence, in addition to a first indicator K1 that considers the strength of the connection. This allows for a more appropriate scoring of the preferential treatment level N1, which in turn is advantageous in selecting preferential customers Us more appropriately.

[0343] Furthermore, as illustrated in Figure 14, the preferential treatment determination system S refers to the passenger's media exposure history and social media activity when calculating the second indicator N2. This allows for a more appropriate reflection of the passenger U's own "strength of social influence" when calculating the second indicator N2. This is advantageous in more appropriately selecting preferential customers Us.

[0344] Furthermore, as illustrated in Figures 17 and 19, the preferential treatment determination system S provides benefits corresponding to the preferential treatment level N1. This encourages PR by preferred customers Us and promotes an improvement in the preferential treatment level N1 for each passenger U. By promoting an improvement in the preferential treatment level N1, it is possible to strengthen the connection between each passenger U and the vehicle V or manufacturer brand through various means such as promoting the use of various services, encouraging visits to dealerships, and strengthening SNS activities.

[0345] Furthermore, as illustrated in Figures 18 and 19, the preferential treatment determination system S grants digital tokens To to preferred customers Us. This further encourages PR by preferred customers Us and improves the preferential treatment level N1 for each passenger U.

[0346] Furthermore, as illustrated in Figures 17 and 19, the preferential treatment determination system S selects passengers U who deserve special treatment as influencers Ui. This allows manufacturers to more appropriately select passengers U who should be sought for cooperation during their PR activities.

[0347] Furthermore, in relation to the estimation of confidence level C1, for example, if the frequency of modifications to the proposal from the proposal unit 112 is high, it can be assumed that passenger U does not trust the proposal. Therefore, as illustrated in graph G2 of Figure 5, it is permissible to estimate passenger U's confidence level C1 lower. By referring to the modification frequency, passenger U's confidence level C1 can be estimated more appropriately.

[0348] Similarly, in relation to the estimation of confidence level C1, for example, if the frequency of accepting proposals from the proposal unit 112 is high, it can be assumed that passenger U trusts the proposals. Therefore, as illustrated in graph G3 of Figure 5, it is permissible to overestimate passenger U's confidence level C1. By referring to the frequency of acceptance, passenger U's confidence level C1 can be estimated more appropriately.

[0349] Similarly, in relation to the estimation of confidence level C1, for example, if the proposal from proposal unit 112 receives a high rating, it can be assumed that passenger U trusts the proposal. Therefore, as illustrated in graph G4 of Figure 5, it is permissible to overestimate passenger U's confidence level C1. By referring to the rating, passenger U's confidence level C1 can be estimated more appropriately.

[0350] Similarly, in relation to the estimation of confidence level C1, for example, when the cumulative operating period of terminal 11 is long, it can be assumed that passenger U trusts the proposal. Therefore, as illustrated in graph G5 of Figure 5, it is permissible to estimate passenger U's confidence level C1 higher. By referring to the cumulative operating period, passenger U's confidence level C1 can be estimated more appropriately.

[0351] [E. Other Embodiments] In the above embodiment, a vehicle V was used as the moving body, but the moving body according to this disclosure is not limited to a vehicle V. The moving body may be a vessel, airplane, or other vehicle other than a vehicle V.

[0352] Furthermore, in the above embodiment, the first indicator K1, the second indicator K2, the sales promotion degree N2, and the preferential treatment degree N1 were calculated by a preferential treatment determination device 1002 located outside the vehicle V, but this disclosure is not limited to such a configuration.

[0353] For example, at least a part of the preferential treatment determination device 1002 may be placed inside the vehicle V. In this case, some or all of the first indicator K1, the second indicator K2, the sales promotion degree N2, and the preferential treatment degree N1 will be calculated inside the vehicle V. [Explanation of Symbols]

[0354] S Preferential Treatment Determination System 1001 Proposal System 1. In-vehicle systems 10 In-Car Devices 11 devices 12. Vehicle control devices 121 Lighting equipment 122 Air conditioning equipment 123 Audio equipment 13 Biometric Information Detection Devices 14. Movement information detection device (onboard equipment) 100 In-vehicle computers 101 Control Unit 111 Estimation Department 112 Proposal Department (Service Provision Department) 112A Action Proposal Department (Proposal Department) 112B Operation proposal section (proposal section) 200 In-vehicle database (storage unit) 300 External Server (Storage Unit) 301 Destination Database 302 Acoustic Database A1 First Questionnaire (Questionnaire) A2 Second Questionnaire (Questionnaire) C0 Personal characteristics C1 Confidence C2 Level of Interest C3 Uniqueness C4 Purpose of boarding C5 Mental and Physical Condition C6 Boarding Risks C7 Preference Characteristics I1 Dialogue Content I2 Biological Information I3 Travel Information J1 Proposal details J2 Presentation method J3 Motivational Information 1002 Preferential Treatment Determination Device 501 Control Unit 514 Communications Department 521 Reliability Acquisition Unit 522 Boarding History Acquisition Section 523 Relationship Acquisition Unit 524 First Calculation Department 531 Influence Acquisition Department 532 Second Calculation Department 541 Preferential Treatment Determination Department 542 Candidate Selection Department 543 Special Features Setting Department 544 Token Granting Section Ne2 Internet L1 confidence data L2 boarding history data L3 Relationship Information L4 Influence Information K1 1st indicator K2 2nd indicator N1 Preferential Treatment N2 Sales promotion level N3 Preferred Customer List N4 Reward List To digital tokens U Passenger US Preferred Customer UI Influencer V Vehicle (mobile object)

Claims

1. A preferential treatment determination system that analyzes multiple passengers boarding each mobile vehicle, A service provision unit mounted on the mobile body and providing services to the passengers, An estimation unit that estimates the degree of trust between the passenger and the mobile vehicle, the service, or the manufacturing brand of the mobile vehicle, based on the passenger's response to the service provided by the service provision unit, A first calculation unit calculates a value of a first indicator that is defined to increase or decrease according to the confidence level estimated by the estimation unit, A preferential treatment determination unit scores the degree of preferential treatment for a plurality of passengers, based on the value of the first indicator calculated by the first calculation unit, for the passengers who were the target of estimation for the first indicator. The system includes a candidate selection unit that selects and outputs passengers who are eligible for preferential treatment from among a plurality of passengers based on the level of the preferential treatment score determined by the preferential treatment determination unit. A system for determining preferential treatment.

2. In the preferential treatment determination system described in claim 1, The mobile body includes a passenger history acquisition unit that acquires the passenger's boarding history based on electrical signals input from the onboard equipment of the mobile body, The first calculation unit calculates the value of the first indicator by referring to the reliability and the boarding history acquired by the boarding history acquisition unit, so as to increase or decrease according to the reliability and the boarding history. A system for determining preferential treatment.

3. In the preferential treatment determination system described in claim 2, The first calculation unit calculates the value of the first indicator by referring to at least one of the following as the passenger's boarding history: the passenger's boarding frequency, the passenger's cumulative boarding time, and the passenger's cumulative distance traveled by the vehicle. A system for determining preferential treatment.

4. In the preferential treatment determination system described in claim 1, A communications unit connected to the internet, The system includes a relationship acquisition unit that acquires relationship information characterizing the depth of the relationship between the passenger and the manufacturing brand via the aforementioned communication unit, The first calculation unit calculates the value of the first index by referring to the confidence level and the relationship information obtained by the relationship acquisition unit, so as to increase or decrease according to the confidence level and the relationship information. A system for determining preferential treatment.

5. In the preferential treatment determination system described in claim 4, The first calculation unit calculates the value of the first indicator by referring to at least one of the following as relational information: information on the passenger's visit to the mobile vehicle's sales outlet, information on visits to the website related to the manufacturing brand, participation history in official events related to the manufacturing brand, and SNS information related to the manufacturing brand. A system for determining preferential treatment.

6. In the preferential treatment determination system described in claim 1, A communications unit connected to the internet, An influence acquisition unit that acquires influence information characterizing the social influence of the passenger via the aforementioned communication unit, The system comprises a second calculation unit that calculates a second indicator, which is defined to increase or decrease according to the strength of the social influence, based on the influence information acquired by the influence acquisition unit, The preferential treatment determination unit scores the preferential treatment so that it increases or decreases according to both the first indicator and the second indicator. A system for determining preferential treatment.

7. In the preferential treatment determination system described in claim 6, The second calculation unit calculates the second indicator by referring to at least one of the following as influence information: the history of exposure in mass media and the status of activity on social media. The social media activity status includes at least one of the following: the passenger's posting history, the number of followers, the number of impressions, and whether or not monetization is possible on the social media platform. A system for determining preferential treatment.

8. In the preferential treatment determination system described in claim 1, The system includes a benefit setting unit that sets benefits to be granted to multiple passengers according to the level of the aforementioned preferential treatment value, The aforementioned benefit setting unit grants to the passenger designated as eligible for preferential treatment at least one of the following: the benefit of lending the mobile device; the benefit of participating in official events related to the manufacturing brand of the mobile device; the benefit of granting exclusive functions to the mobile device; and a monetary benefit when purchasing or selling the mobile device. A system for determining preferential treatment.

9. In the preferential treatment determination system described in claim 8, The system includes a token granting unit that grants digital tokens managed on the blockchain to the passengers designated as eligible for the preferential treatment. The value of the aforementioned digital token fluctuates according to the value of the first indicator or the preferential treatment level. The aforementioned benefit setting unit sets different benefits to passengers designated as eligible for preferential treatment according to the value of the digital token. A system for determining preferential treatment.

10. In the preferential treatment determination system described in claim 1, The candidate selection department shall designate as influencers passengers who, among those designated as eligible for preferential treatment, have a preferential treatment level equal to or greater than a predetermined threshold. A system for determining preferential treatment.

11. In the preferential treatment determination system described in claim 1, The mobile device is equipped with an interactive terminal, The terminal accepts the passenger's request for modifications to the content of the service provided. The estimation unit calculates the frequency or amount of modification of the content of the service provided via the terminal, The estimation unit updates the reliability according to the correction frequency or correction amount. A system for determining preferential treatment.

12. In the preferential treatment determination system described in claim 1, The mobile device is equipped with an interactive terminal, The terminal receives confirmation from the passenger whether or not they will accept the service. The estimation unit calculates the frequency of adoption of the provided content via the terminal, The estimation unit updates the reliability according to the frequency of adoption. A system for determining preferential treatment.

13. In the preferential treatment determination system described in claim 1, The mobile device is equipped with an interactive terminal, The terminal receives the passenger's evaluation of the service provided. The estimation unit aggregates the evaluations received by the terminal, The estimation unit updates the confidence level according to the aggregated results of the evaluation. A system for determining preferential treatment.

14. In the preferential treatment determination system described in claim 1, The mobile device is equipped with an interactive terminal, The estimation unit determines the cumulative operating period of the terminal, The estimation unit updates the reliability according to the cumulative operating period. A system for determining preferential treatment.

Images