Information processing device and information processing method

Abstract

The information processing device (100) includes: a controller (60); a vehicle boarding detection device (sensor group 50) that detects the boarding of a user (80); a storage device (53) that records the behavior data of the user (80); an output device (speaker 54) that outputs question data to be answered by the user (80); and an input device (microphone 52) that receives input from the user (80). When the controller (60) detects the boarding of the user (80) based on a signal obtained from the vehicle boarding detection device, the controller (60) outputs output data including at least a question related to the behavior of the user (80) before boarding the vehicle based on the position information of the vehicle (40), obtains the user (80)'s answer to the question as input data via the input device, and stores the input data in the storage device (53) in association with the position information of the vehicle (40) or a point of interest.

Description

Technical Field

[0001] The present invention relates to an information processing device and an information processing method. Background Art

[0002] Conventionally, a navigation device that displays icons of POIs (Points of Interest) is known (Patent Document 1). In the invention described in Patent Document 1, the frequency of use of POIs used by a user is recorded, and icons of POIs with high frequency of use are displayed on a display.

[0003] Prior art literature

[0004] Patent Literature

[0005] Patent Document 1: Japanese Patent Application Laid-Open No. 2012-57957 Summary of the Invention

[0006] Problems to be solved by the invention

[0007] However, the invention described in Patent Document 1 merely displays frequently used POIs on a display and does not mention recording user behavior at the POIs. Therefore, there is room for improvement.

[0008] The present invention has been made in view of the above-mentioned problems, and an object of the present invention is to provide an information processing device and an information processing method capable of recording a user's behavior at a POI.

[0009] Solutions for solving problems

[0010] An information processing device involved in one embodiment of the present invention, when detecting a user's boarding of a vehicle based on a signal obtained from a boarding detection device, outputs output data containing at least questions related to the user's pre-boarding behavior from an output device based on the vehicle's location information, obtains the user's answers to the questions as input data via an input device, and stores the input data in a storage device in association with the vehicle's location information or POI.

[0011] Effects of the Invention

[0012] According to the present invention, the user's actions at a POI can be recorded. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 It is a diagram for explaining the overall outline of the embodiment of the present invention.

[0014] Figure 2 This is a schematic configuration diagram of an information processing device according to an embodiment of the present invention.

[0015] Figure 3This is a diagram illustrating an example of POI according to the embodiment of the present invention.

[0016] Figure 4 It is a diagram for explaining another example of POI according to the embodiment of the present invention.

[0017] Figure 5 This is a flowchart illustrating an operation example of the information processing device according to the embodiment of the present invention.

[0018] Figure 6 It is a schematic configuration diagram of an information processing device according to a modified example of the present invention.

[0019] Figure 7 This is a diagram illustrating the accuracy of estimated behavior according to a modified example of the present invention.

[0020] Figure 8 This is a diagram illustrating the structure of questions according to a modified example of the present invention.

[0021] Figure 9 This is a diagram illustrating the relationship between action history records and the accuracy of estimated actions.

[0022] Figure 10 This is a flowchart illustrating an operation example of the information processing device according to the modified example of the present invention.

[0023] Figure 11 It is a diagram for explaining another embodiment of the present invention. DETAILED DESCRIPTION

[0024] Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. In the description of the drawings, the same parts are denoted by the same reference numerals and their description will be omitted.

[0025] (Overall overview diagram)

[0026] Reference Figure 1 The overall outline of this embodiment is described below. Figure 1 As shown, vehicle 40 communicates with computer 20 via communication network 30 .

[0027] The computer 20 includes a CPU (Central Processing Unit) 21, a memory 22, a communication I / F 23, and a storage device 24. These components are electrically connected via a bus (not shown). The installation location of the computer 20 is not particularly limited and can be installed at any location.

[0028] CPU 21 reads various programs stored in storage device 24 and other devices into memory 22 and executes the various commands contained in the programs. Memory 22 is a storage medium such as ROM (Read Only Memory) or RAM (Random Access Memory). Storage device 24 is a storage medium such as an HDD (Hard Disk Drive). Furthermore, the functions of computer 20 can also be provided through application programs (such as Software as a Service (SaaS)) deployed on communication network 30. Alternatively, computer 20 can be a server.

[0029] The communication I / F 23 is implemented as hardware such as a network adapter, various communication software, or a combination thereof, and is configured to enable wired or wireless communication via the communication network 30 or the like.

[0030] The communication network 30 may be configured by either wireless or wired means, or both, and may also include the Internet. In this embodiment, the computer 20 and the vehicle 40 are connected to the communication network 30 by wireless communication.

[0031] (Configuration Example of Information Processing Device)

[0032] Next, refer to Figure 2 Next, a configuration example of the information processing device 100 mounted on the vehicle 40 will be described.

[0033] like Figure 2 As shown, the information processing device 100 includes a sensor group 50 , a GPS receiver 51 , a microphone 52 , a storage device 53 , a speaker 54 , and a controller 60 .

[0034] The information processing device 100 can be installed in a vehicle with an automatic driving function or in a vehicle without an automatic driving function. In addition, the information processing device 100 can also be installed in a vehicle that can switch between automatic driving and manual driving. In addition, the automatic driving in this embodiment refers to a state in which at least any one of the actuators, such as the brake actuator, the accelerator actuator, and the steering wheel actuator, is controlled without user operation. Therefore, it is also possible for other actuators to work through user operation. In addition, automatic driving refers to a state in which any one of the acceleration and deceleration control, lateral position control, etc. is being automatically performed. In addition, manual driving in this embodiment refers to a state in which the user operates the brake pedal, the accelerator pedal, and the steering wheel, for example.

[0035] The sensor group 50 (occupancy detection device) is used to detect when a user is boarding a vehicle. It includes a pressure sensor (also called a seating sensor) mounted on the seat cushion, a camera that captures images of the user inside the vehicle, and sensors that detect door opening and closing. The controller 60 uses signals acquired from these sensors to detect when a user is boarding a vehicle.

[0036] The GPS receiver 51 detects the position information of the vehicle 40 on the ground by receiving radio waves from artificial satellites. The position information of the vehicle 40 detected by the GPS receiver 51 includes latitude and longitude information. The GPS receiver 51 outputs the detected position information of the vehicle 40 to the controller 60. The method for detecting the position information of the vehicle 40 is not limited to using the GPS receiver 51. For example, the position can also be estimated using a method known as odometry. Odometry is a method of estimating the position of the vehicle 40 by calculating the amount and direction of movement of the vehicle 40 based on the rotation angle and angular velocity of the vehicle 40.

[0037] The microphone 52 is used to input the user's voice.

[0038] The storage device 53 is a storage device different from the memory of the controller 60 and is, for example, a hard disk, a solid state drive, etc. The storage device 53 stores a map database 53a, an action history database 53b, and a question database 53c.

[0039] The map database 53a stores map information required for route guidance, such as road information and facility information. Road information is, for example, information related to the number of lanes on the road, road dividing lines, and the connection relationship between lanes. The map database 53a outputs map information to the controller 60 in response to a request from the controller 60. In this embodiment, it is assumed that the information processing device 100 has a map database 53a for illustration, but the information processing device 100 does not necessarily have a map database 53a. It is also possible to obtain map information by using vehicle-to-vehicle communication or road-to-vehicle communication. In addition, when the map information is stored on an external server (e.g. Figure 1 In the case of the computer 20 shown in FIG. 1 , the information processing device 100 may also obtain map information from the server at any time through communication. In addition, the information processing device 100 may also periodically obtain the latest map information from the server and update the map information it holds.

[0040] The map database 53a also stores information related to POIs (POINT OF INTEREST). In this embodiment, a POI is data representing a specific location. A POI includes at least attributes and location information (latitude and longitude). Attributes refer to information used to classify POIs, such as restaurants, shopping malls, and parks. Furthermore, a POI may also include a name, address, phone number, and icon. For example, if the attribute of a POI is a restaurant, the name refers to the specific store name of the restaurant. In addition to latitude and longitude, the location information of a POI may also include altitude.

[0041] The action history database 53b stores the user's actions at the POI in association with the position information or the POI of the vehicle 40. Thus, the user's actions at the POI are stored as history records in the database.

[0042] In the question database 53c, attributes of POIs and questions related to user actions at the POIs are recorded in association with each other.

[0043] The controller 60 is a general-purpose microcomputer having a CPU (central processing unit), a memory, and an input / output unit. A computer program for making it function as an information processing device 100 is installed in the microcomputer. The microcomputer functions as a plurality of information processing circuits possessed by the information processing device 100 by executing the computer program. In addition, an example of realizing the plurality of information processing circuits possessed by the information processing device 100 by software is shown here, but of course it is also possible to prepare dedicated hardware for performing each information processing shown below to constitute the information processing circuit. In addition, it is also possible to constitute a plurality of information processing circuits by separate hardware. The controller 60 has a user determination unit 61, a destination arrival judgment unit 62, a sound analysis unit 63, an action history update unit 64, an action estimation unit 65, a question selection unit 66, a vehicle riding detection unit 67, and a question output unit 68 as an example of a plurality of information processing circuits.

[0044] The speaker 54 is provided in the cabin of the vehicle 40 and is used to output sound.

[0045] The user identification unit 61 identifies the user using a facial image captured while the user is boarding the vehicle 40. Specifically, the user identification unit 61 determines whether the captured facial image matches or is similar to a facial image pre-registered in the storage device 53. If the captured facial image matches or is similar to a facial image pre-registered in the storage device 53, the user identification unit 61 identifies the user boarding the vehicle 40 as a pre-registered user. Furthermore, this facial recognition can be performed even when the ignition is off.

[0046] As another method of identifying the user, the ID of an intelligence key (sometimes also called a smart key) can also be used. In a smart key system, an antenna that transmits radio waves and a receiver that receives radio waves are installed on both the vehicle and the key. When the user presses a switch attached to a door handle, trunk, etc., radio waves are transmitted from the vehicle's antenna, and the key, upon receiving the radio waves, automatically transmits radio waves back. The vehicle's receiver receives these radio waves to lock or unlock the vehicle. The radio waves transmitted by the key contain an identification key, which is pre-registered in the vehicle for use. By associating this identification key with user information, the user can be identified.

[0047] The destination arrival determination unit 62 determines whether the vehicle 40 has arrived at the destination. In this embodiment, the destination is the destination set by the user using a navigation device (not shown). The location information of the destination is stored in the map database 53a. The destination arrival determination unit 62 compares the location information of the vehicle 40 obtained from the GPS receiver 51 with the location information of the destination stored in the map database 53a. If the two match or approximately match, the vehicle 40 determines that the destination has arrived.

[0048] The voice analysis unit 63 analyzes the user's voice input via the microphone 52. A well-known method is used as the analysis method.

[0049] The action history updating unit 64 stores the action performed by the user at the POI in association with the position information of the vehicle 40 or the POI in the action history database 53 b , and updates the action history of the user at the POI.

[0050] The behavior estimation unit 65 estimates the user's pre-boarding behavior. In this embodiment, the user's pre-boarding behavior refers to the user's behavior before boarding the vehicle 40. More specifically, the user's pre-boarding behavior refers to the user's behavior at the destination before boarding the vehicle 40.

[0051] Furthermore, the activity estimation unit 65 obtains the user's schedule data from the computer 20 via the communication network 30. The user's schedule data includes information related to the user's planned activities, such as when, where, and what. Alternatively, the schedule data may be obtained from a terminal held by the user (e.g., a smartphone).

[0052] The question selection unit 66 selects a question to be asked to the user by referring to the behavior estimated by the behavior estimation unit 65 and the question database 53c. When the question selection unit 66 selects a question, a signal indicating the selected question is output to the question output unit 68.

[0053] The boarding detection unit 67 detects the user's boarding based on the signals received from the sensor group 50. Specifically, the boarding detection unit 67 detects the user's boarding when it receives a signal indicating a change in resistance from the pressure sensor. Alternatively, the boarding detection unit 67 detects the user's boarding when it analyzes the camera image and detects the user's presence inside the vehicle. Alternatively, the boarding detection unit 67 may detect the user's boarding when it receives a signal indicating a change in resistance from the pressure sensor after detecting the opening and closing of a door. When the boarding detection unit 67 detects the user's boarding, a signal indicating the user's boarding is output to the question output unit 68.

[0054] Upon receiving a signal indicating that user 80 has been detected boarding a vehicle, question output unit 68 outputs the question selected by question selection unit 66. Questions may be outputted audibly via speaker 54 or as text messages on a display (e.g., a navigation device display). In this embodiment, the question is described as being outputted audibly via speaker 54.

[0055] Next, refer to Figure 3 An example of a method for storing actions performed by a user at a POI will be described.

[0056] Figure 3 The illustrated scene shows a scene where vehicle 40 arrives at a destination set by user 80 (here, ramen restaurant 90) and parks in the parking lot of ramen restaurant 90. User 80 gets out of vehicle 40 and enters ramen restaurant 90. After completing their stay at ramen restaurant 90, user 80 exits ramen restaurant 90 and gets back into vehicle 40.

[0057] When the destination arrival determination unit 62 determines that the vehicle 40 has arrived at the destination, a signal indicating that the vehicle 40 has arrived at the destination is output to the behavior estimation unit 65. The behavior estimation unit 65, which receives the signal, estimates the behavior of the user 80 before boarding the vehicle. As mentioned above, the behavior of the user 80 before boarding the vehicle refers to the behavior of the user 80 at the destination when the user 80 is not boarding the vehicle 40. Figure 3 In the example shown, the destination of the user 80 is the ramen restaurant 90 , and therefore the actions of the user 80 before boarding the bus refer to the actions performed by the user 80 at the ramen restaurant 90 .

[0058] The following describes a method by which the behavior estimation unit 65 estimates the behavior of the user 80. First, the behavior estimation unit 65 obtains the position information of the vehicle 40 from the GPS receiver 51. More specifically, the behavior estimation unit 65 obtains the position information of the vehicle 40 when the destination arrival determination unit 62 determines that the vehicle 40 has arrived at the destination.

[0059] The behavior estimation unit 65 compares the position information of the vehicle 40 with the map database 53a to obtain the POI of the current location (the parking location of the vehicle 40). Figure 3 As shown in FIG, the action estimation unit 65 obtains the position information (latitude, longitude) of the POI and the attribute of the POI (ramen shop). Figure 3 In the example shown, vehicle 40 is parked in the parking lot of ramen restaurant 90. Therefore, the location information of vehicle 40 is considered to be the location information of the POI. The behavior estimation unit 65 obtains the location information of the POI that matches or substantially matches the location information of vehicle 40. The behavior estimation unit 65 also obtains the attributes of the POI associated with the location information of the POI.

[0060] In this way, the action estimation unit 65 determines that the current location is the ramen restaurant 90 by comparing the location information of the vehicle 40 with the map database 53a. Next, the action estimation unit 65 estimates the action of the user 80 before boarding the vehicle by referring to the table data stored in the storage device 53, which associates the attributes of the POI with the user's action type. In this embodiment, the user's action type refers to the category into which the actions that the user may perform at the POI are classified. For example, in the case where the attribute of the POI is a restaurant, the two actions of dining and discussing are stored as the user's action types in association with the restaurant. However, the user's action type may be a single one instead of multiple. For example Figure 3 As shown in FIG, in the case where the attribute of the POI is a ramen restaurant, only dining is associated with the ramen restaurant as the user's action type. Figure 3 In the example shown, the activity of the user 80 before boarding the vehicle estimated by the activity estimation unit 65 is eating.

[0061] A signal related to the estimation result obtained by the behavior estimation unit 65 is output to the question selection unit 66. Upon receiving the signal from the behavior estimation unit 65, the question selection unit 66 selects a question to be asked to the user by referring to the user 80's pre-boarding behavior (eating) and the question database 53c. As described above, the question database 53c stores POI attributes in association with questions related to the user's behavior at the POI. By referring to the user 80's pre-boarding behavior (eating) and the question database 53c, the question selection unit 66 can select a question related to eating. For example, stored questions related to eating include "Have you eaten?", which can be answered with either "yes" or "no." Questions requiring a specific answer, such as "What did you eat?", are also stored. The question selection unit 66 can select any question related to eating. When the question selection unit 66 selects a question, a signal indicating the selected question is output to the question output unit 68.

[0062] The timing for asking the question to user 80 is not particularly limited, but it can be considered to be immediately after user 80 boards vehicle 40. This is because as time passes since user 80 boarded vehicle 40, user 80 may forget the actions taken at the POI. Therefore, the boarding detection unit 67 detects the user's boarding based on the signals acquired from the sensor group 50. When the boarding detection unit 67 detects the boarding of user 80, a signal indicating the detection of the boarding of user 80 is output to the question output unit 68.

[0063] When receiving a signal indicating that the user 80 has been detected boarding the vehicle, the question output unit 68 asks the user the question selected by the question selection unit 66 via the speaker 54. Here, it is assumed that the question "Have you eaten?" is asked.

[0064] User 80 answers this question using microphone 52. Voice analysis unit 63 analyzes user 80's voice data. If user 80 answers "yes," voice analysis determines that user 80's action at ramen restaurant 90 was to eat. Action history update unit 64 associates the vehicle 40's location information with user 80's action of eating at ramen restaurant 90 and stores it in action history database 53b. Alternatively, action history update unit 64 may associate the POI with user 80's action of eating at ramen restaurant 90 and store it in action history database 53b. Thus, according to this embodiment, user 80's actions at POIs can be recorded.

[0065] The data stored in the action history database 53 b include, for example, the date and time when the user 80 inputs the answer, the position information of the vehicle 40 , the POI (position information and attributes), and the content of the action at the POI.

[0066] Next, refer to Figure 4 Another example of a method of storing actions performed by a user at a POI will be described.

[0067] Figure 4 The illustrated scene shows a vehicle 40 arriving at a destination set by a user 80 (here, a shopping mall 91) and parking in the parking lot of the shopping mall 91. The user 80 gets out of the vehicle 40 and enters the shopping mall 91. After the user 80 completes their business at the shopping mall 91, they exit the shopping mall 91 and board the vehicle 40.

[0068] and Figure 3Similarly to the example shown, the behavior estimation unit 65 compares the position information of the vehicle 40 with the map database 53a to determine that the current location is the shopping mall 91. Next, the behavior estimation unit 65 estimates the behavior of the user 80 by referring to the table data associating POI attributes with the user's behavior type.

[0069] In this embodiment, the shopping mall 91 is a commercial facility that houses a plurality of retail stores, restaurants, beauty salons, travel agencies, gyms, etc. In addition, a shopping mall is sometimes also called a shopping center.

[0070] The types of user activities in shopping mall 91 include dining, shopping, dating, getting a haircut, exercising, and many other activities. To accurately estimate user 80's activities from these multiple activities, activity estimation unit 65 refers to user 80's schedule data acquired from computer 20. As mentioned above, schedule data contains information regarding user 80's planned activities. Therefore, by referring to schedule data, activity estimation unit 65 can accurately estimate the user's activities in shopping mall 91. Here, assume that the schedule data includes information that user 80 is dining in shopping mall 91 as his planned activity. In this case, activity estimation unit 65, by referring to the schedule data, estimates that the user's activities in shopping mall 91 are dining.

[0071] The question selection unit 66, which receives the signal from the behavior estimation unit 65, selects questions to be asked to the user by referring to the user 80's behavior (eating) before boarding the bus and the question database 53c. Figure 3 In the example shown, it is assumed that the question “Have you eaten?” is asked.

[0072] In response to this question, the user 80 answers the question by voice using the microphone 52. The voice analysis unit 63 analyzes the voice data of the user 80. If the user 80 answers "yes", the voice analysis determines that the action performed by the user 80 in the shopping plaza 91 is to have a meal. The action history update unit 64 associates the location information of the vehicle 40 with the action of the user 80 having a meal in the shopping plaza 91 and stores it in the action history database 53b. In addition, the action history update unit 64 may also associate the POI with the action of the user 80 having a meal in the shopping plaza 91 and store it in the action history database 53b. In this way, even if the user has multiple types of actions at the POI, the actions of the user 80 at the POI can be accurately recorded by referring to the schedule data.

[0073] Alternatively, the action estimation unit 65 may estimate the action of the user 80 by referring to the action history of the user 80 stored in the action history database 53b. For example, the action with the highest frequency among the actions stored in the action history database 53b may be estimated as the action of the user 80.

[0074] Next, refer to Figure 5 An operation example of the information processing device 100 is described with reference to the flowchart of FIG.

[0075] In step S101, the user identification unit 61 identifies the user 80 boarding the vehicle 40 using a facial image captured by a camera, an identification key transmitted from a smart key, etc. The process proceeds to step S103, where the user 80 sets a destination.

[0076] The process proceeds to step S105, where the destination arrival determination unit 62 compares the vehicle 40's location information acquired from the GPS receiver 51 with the destination's location information stored in the map database 53a to determine whether the vehicle 40 has arrived at the destination. If the vehicle 40 is determined to have arrived at the destination (step S105: "Yes"), the process proceeds to step S107. On the other hand, if the vehicle 40 is not determined to have arrived at the destination (step S105: "No"), the process enters a standby state.

[0077] In step S107, the behavior estimation unit 65 estimates the user's behavior before boarding the vehicle. Specifically, the behavior estimation unit 65 compares the location information of the vehicle 40 obtained from the GPS receiver 51 with the map database 53a to obtain the POI of the current location (the parking location of the vehicle 40). In this way, the behavior estimation unit 65 can understand that the current location is the ramen restaurant 90 (refer to Figure 3 ) or the current location is Shopping Plaza 91 (refer to Figure 4 The behavior estimation unit 65 estimates the pre-boarding behavior of the user 80 by referring to table data that associates the attributes of the POI (ramen shop, shopping mall) with the user's behavior type. A signal related to the estimation result obtained by the behavior estimation unit 65 is output to the question selection unit 66.

[0078] The process proceeds to step S109 , where the question selection unit 66 , having received the signal from the behavior estimation unit 65 , selects a question to be asked to the user by referring to the user 80's pre-boarding behavior and the question database 53 c. A signal indicating the question selected by the question selection unit 66 is output to the question output unit 68 .

[0079] The process proceeds to step S111, where the boarding detection unit 67 detects the user boarding based on the signal acquired from the sensor group 50. When the boarding detection unit 67 detects the user 80 boarding, a signal indicating that the user 80 boarding is detected is output to the question output unit 68.

[0080] The process proceeds to step S113 , and upon receiving the signal indicating that the user 80 has been detected boarding the vehicle, the question output unit 68 voice-interprets the question selected by the question selection unit 66 to the user via the speaker 54 .

[0081] If a voice response is received via microphone 52 (step S115: Yes), the process proceeds to step S117. If no response is received from user 80 (step S115: No), the process remains on standby. The method of receiving a response from user 80 is not limited to voice; the response from user 80 may also be received through touch panel operation.

[0082] In step S117, the voice analysis unit 63 analyzes the voice data of the user 80. Based on the analysis results obtained by the voice analysis unit 63, the action history update unit 64 associates the location information of the vehicle 40 with the actions of the user 80 at the POI and stores them in the action history database 53b. The action history update unit 64 may also associate the POI with the actions of the user 80 at the POI and store them in the action history database 53b.

[0083] (Effect)

[0084] As described above, according to the information processing device 100 according to this embodiment, the following operational effects can be obtained.

[0085] The information processing device 100 includes: a controller 60; a riding detection device (sensor group 50) that detects the user 80 riding in the vehicle 40; a storage device (action history database 53b) that records the action data of the user 80; an output device (speaker 54) that outputs question data requiring the user 80 to answer; and an input device (microphone 52) that accepts input from the user 80.

[0086] When the controller 60 detects the boarding of the vehicle 80 based on the signal received from the boarding detection device, it outputs output data from the output device, including at least questions regarding the user 80's actions before boarding the vehicle, based on the vehicle 40's location information. The controller 60 receives the user 80's responses to the questions as input data via the input device. The controller 60 then stores the input data in the storage device, in association with the vehicle 40's location information or a point of interest (POI). This allows the user 80's actions at the POI to be accurately recorded.

[0087] The controller 60 determines the question to be asked based on the attributes of the POI determined based on the position information of the vehicle 40. Figure 3 As shown, vehicle 40 has parked in the parking lot of ramen restaurant 90, so the POI attribute is determined to be ramen restaurant 90 based on the location information of vehicle 40. The storage device (question database 53c) records the POI attribute (ramen restaurant) in association with questions related to the user's behavior (eating) at the POI, so that controller 60 can determine questions based on the POI attribute.

[0088] The controller 60 may also estimate the behavior before boarding the vehicle based on the behavior history or the attributes of the POI stored in the behavior history database 53b. Figure 3 As shown, if the POI attribute is a ramen restaurant, the controller 60 can estimate the pre-boarding activity as a meal. Alternatively, the controller 60 can estimate the most frequent activity among the activities stored in the activity history database 53b as the activity of the user 80. This allows the controller 60 to accurately estimate the pre-boarding activity.

[0089] Controller 60 can also acquire data related to user 80's planned activities (user 80's schedule data) through communication and determine questions based on this schedule data. This allows appropriate questions to be determined even if the user has multiple types of activities at a POI, accurately recording user 80's activities at the POI. Communication refers to communication with computer 20 (server) or with a terminal held by user 80.

[0090] The input data is data converted from the user 80's voice or data generated by receiving an operation from the user 80. A microphone 52 and a touch panel are used as answering means for the user 80 to answer questions. The user 80 can easily answer questions using the microphone 52 and the touch panel.

[0091] The controller 60 detects the user 80 getting off the vehicle 40 after the vehicle 40 arrives at the destination set by the user 80, and outputs a question when the controller 60 detects the user 80 getting on the vehicle thereafter.

[0092] The POI may also be any one of a POI located around the location information of the vehicle 40, a destination set in the navigation device of the vehicle 40, a POI included in data related to the user's 80 action plan obtained through communication, and a POI included in data related to the user's 80 action history before boarding the vehicle obtained through communication.

[0093] [Modification]

[0094] Next, a modification of this embodiment will be described.

[0095] like Figure 6 As shown, the information processing device 101 according to the modified example further includes an accuracy calculation unit 69. The accuracy calculation unit 69 calculates the accuracy indicating the accuracy (probability) of the behavior of the user 80 estimated by the behavior estimation unit 65. The estimation accuracy can be evaluated in three levels: low, medium, and high, or as a probability (0% to 100%).

[0096] Reference Figure 7 An example of a method for storing actions performed by the user 80 at a POI will be described.

[0097] exist Figure 7 In this example, restaurants, shopping malls, and classrooms are listed as POI attributes. In this embodiment, classrooms include swimming classes, dance classes, and flower arrangement classes. User activities in restaurants include dining and discussion. As mentioned above, user activities in shopping malls include dining, shopping, dating, haircuts, and exercise. User activities in classrooms include exercise and practice.

[0098] As described above, the attributes of POI are associated with the type of user's behavior, so Figure 7 As shown, the action estimation unit 65 estimates the action of the user 80 from the above-mentioned action types. Figure 7 In the example, it is shown that the user's estimated activities in the shopping mall are dining, shopping, and dating.

[0099] like Figure 7 As shown, when there are multiple actions to be estimated, the action estimated by the action estimation unit 65 (hereinafter referred to as the estimated action) is required to be highly accurate, but the estimation is not always performed with high accuracy. The reason for requiring high accuracy in the estimated action is that if the estimated action is incorrect, the question to the user 80 will deviate from the main point. However, as mentioned above, the estimation is not always performed with high accuracy.

[0100] Therefore, the information processing device 101 according to the modified example calculates the accuracy of the estimated behavior and determines the question based on the calculated accuracy. The accuracy of the estimated behavior is calculated using the action history of user 80 stored in action history database 53b. Here, the following describes two cases: a case where the action history of user 80 stored in action history database 53b is zero, and a case where the action history of user 80 stored in action history database 53b is a predetermined number of times (e.g., 10).

[0101] First, a case where the action history record of the user 80 stored in the action history database 53b is zero will be described.

[0102] In the case where the destination of user 80 is a restaurant, Figure 7 As shown, the activity estimated by the activity estimation unit 65 is either dining or discussing. The activity estimation unit 65 estimates either dining or discussing. The estimation method is not particularly limited, but as an example, the activity estimation unit 65 may estimate an activity that is generally assumed to have a high probability. Among the activities in a restaurant, the activity that is generally assumed to have a high probability is dining. This fact is also stored in the table data. Therefore, when the POI attribute is restaurant, the activity estimation unit 65 may refer to the table data and estimate the user 80's activity as dining.

[0103] As another example, the action estimation unit 65 may refer to the action history of the user 80 stored in the action history database 53b to estimate the action of the user 80. For example, among the actions stored in the action history database 53b, the action with the highest frequency may be estimated as the action of the user 80. However, this method cannot be used if the action history record is zero.

[0104] The accuracy calculation unit 69 calculates the accuracy of the action of dining estimated by the action estimation unit 65. First, the accuracy calculation unit 69 refers to the action history database 53b to obtain the action history of the user 80 in the restaurant. Here, the action history of the user 80 in the restaurant is zero, so the accuracy calculation unit 69 determines that the user 80 is visiting the restaurant for the first time. In this case, the accuracy calculation unit 69 determines that the possibility that the action performed by the user 80 in the restaurant is dining is low. The reason is that it is difficult to determine whether the action performed by the user 80 in the restaurant is dining or discussing. It is assumed that the probability of dining is generally higher, but the possibility of having a discussion cannot be ruled out. Therefore, the accuracy calculation unit 69 calculates the accuracy of the action of dining estimated by the action estimation unit 65 to be low. In addition, a signal indicating the calculation result is output to the question selection unit 66.

[0105] Question selection unit 66, having received the signal from accuracy calculation unit 69, determines that the accuracy of the dining activity estimated by activity estimation unit 65 is low. Based on this determination, question selection unit 66 avoids selecting questions related to dining and selects questions that inquire about the user 80's activity itself. The reason for selecting questions that inquire about the user 80's activity itself is to accumulate activity history in preparation for the user 80's future visit to the restaurant.

[0106] As questions about the action itself, e.g. Figure 7As shown, a question such as "What did you do?" can be asked. User 80's answer to the question is stored in the action history database 53b, associated with the vehicle 40's location information or POI. In this way, if the action history of user 80 stored in action history database 53b is zero, by asking questions about the user's actions themselves, the action history can be accumulated in preparation for the user's future visit to the restaurant. Furthermore, by asking questions about the user's actions themselves, off-topic questions can be avoided.

[0107] Next, a case where the action history records of the user 80 stored in the action history database 53 b are a predetermined number of times (for example, 10 times) will be described.

[0108] Assuming that user 80's action history stored in action history database 53b is zero, assume that user 80's destination is a restaurant. The accuracy calculation unit 69 references action history database 53b to obtain user 80's action history at the restaurant. Here, assume that, out of 10 action history records, user 80 dined seven times and discussed three times. In this case, the number of dined times exceeds the number of discussed times, so the accuracy calculation unit 69 calculates the probability that user 80's action at the restaurant was dining as medium. In other words, the accuracy calculation unit 69 calculates the accuracy of the action estimate unit 65's estimate of the action of dining as medium. A signal indicating the calculation result is output to the question selection unit 66. Alternatively, the accuracy of the action estimate unit 65's estimate of the action of dining can be calculated as a probability. If, out of 10 action history records, user 80 dined seven times and discussed three times, the probability that user 80's action at the restaurant was dining can be calculated as 70%. The medium accuracy of the action estimate unit 65's estimate of the action is approximately 70% when converted into a probability.

[0109] The question selection unit 66, which receives the signal from the accuracy calculation unit 69, determines that the user 80 is likely to have a meal in the restaurant, and selects a question related to the meal. Figure 7 As shown in FIG, questions such as “Have you eaten?” and “Have you drunk?” can be given. The user 80's answers to the questions are stored in the action history database 53b in association with the location information or POI of the vehicle 40. In addition, if there are 3 times of dining and 7 times of discussion in the 10 action history records, the question selection unit 66 selects questions related to discussion. As questions related to discussion, Figure 7 As shown, questions such as "What was discussed?" and "With whom?" can be given as examples.

[0110] While the prescribed number of times is described as 10, it is not limited to 10. The greater the number of action history records of user 80 stored in action history database 53b, the more accurate the estimated action is. In other words, the fewer the number of action history records of user 80 stored in action history database 53b, the more the estimated action accuracy deviates. Therefore, if the number of action history records of user 80 stored in action history database 53b is less than 5, the number of action history records of user 80 stored in action history database 53b may be set to zero for processing.

[0111] Similarly, when the destination of the user 80 is a shopping mall and the accuracy of the user 80's behavior (dining) estimated by the behavior estimation unit 65 is calculated to be medium, the question selection unit 66 selects questions related to dining. Questions related to dining are the same as those described above and are therefore omitted. Similarly, when the accuracy of the user 80's behavior (shopping) estimated by the behavior estimation unit 65 is calculated to be medium, the question selection unit 66 selects questions related to shopping. As questions related to shopping, Figure 7 As shown in FIG, questions such as “Did you buy anything?” and “What do you have?” can be given. Similarly, when the accuracy of the action (date) of the user 80 estimated by the action estimation unit 65 is calculated to be medium, the question selection unit 66 selects questions related to the date. As questions related to the date, Figure 7 As shown, questions such as "Who are you going on a date with?" and "Is it ○○?" can be asked.

[0112] Similarly, when the destination of the user 80 is the classroom and the accuracy of the user 80's behavior (movement) estimated by the behavior estimation unit 65 is calculated to be medium, the question selection unit 66 selects questions related to movement. Figure 7 As shown in FIG, questions such as “Have you exercised?”, “Swimming? Yoga? Dance?” can be given. Similarly, when the accuracy of the action (exercise) of the user 80 estimated by the action estimation unit 65 is calculated to be medium, the question selection unit 66 selects questions related to exercise. As questions related to exercise, Figure 7 As shown, questions such as "Are you learning something?", "Flower arrangement? Tea ceremony? Calligraphy practice?" can be cited.

[0113] In this way, by calculating the accuracy of the behavior of the user 80 estimated by the behavior estimation unit 65 and determining questions based on the calculated accuracy, it is possible to avoid asking questions that deviate from the main point.

[0114] Next, assume that out of 10 history records of user 80's actions at a restaurant, 10 records include dining and 0 records include discussion. In this case, since all of the history records are about dining, accuracy calculation unit 69 calculates the probability that user 80's actions at the restaurant were dining as high. In other words, accuracy calculation unit 69 calculates the accuracy of user 80's actions (eating) estimated by action estimation unit 65 as high. A signal indicating the calculation result is then output to question selection unit 66. Alternatively, the probability that user 80's actions at the restaurant were dining can be calculated as 100%.

[0115] The question selection unit 66, which receives the signal from the accuracy calculation unit 69, determines that the user 80 is highly likely to have a meal in the restaurant, and selects a more in-depth question from the questions related to the meal than the question with a medium accuracy. Figure 7 As shown, questions such as "Is it delicious?" and "Would you like to eat here again?" can be asked. The voice analysis unit 63 can classify the user 80's answers to these questions as affirmative or negative. The action history update unit 64 can record the classification results as classification data in the action history database 53b, in association with the POI. By classifying the user 80's answers as affirmative or negative, more in-depth questions can be asked to the user 80 the next time.

[0116] Furthermore, if there are 0 dining and 10 discussions in the 10 action history records, the question selection unit 66 selects a more in-depth question from the discussion-related questions than the question with a medium accuracy. Figure 7 As shown, questions such as "Did it end smoothly?" and "Is ○○ okay?" can be asked.

[0117] Similarly, when the destination of the user 80 is a shopping mall and the accuracy of the action (dining) of the user 80 estimated by the action estimation unit 65 is calculated to be high, the question selection unit 66 selects questions from questions related to dining that are more in-depth than the questions when the accuracy is medium. The in-depth questions are the same as those described above and are therefore omitted. Similarly, when the accuracy of the action (shopping) of the user 80 estimated by the action estimation unit 65 is calculated to be high, the question selection unit 66 selects questions from questions related to shopping that are more in-depth than the questions when the accuracy is medium. As in-depth questions, such as Figure 7 As shown in FIG, questions such as “Are there any cheap things?” and “Did you buy something good?” can be given. Similarly, when the accuracy of the action (date) of the user 80 estimated by the action estimation unit 65 is calculated to be high, the question selection unit 66 selects questions related to the date that are more in-depth than the questions when the accuracy is medium. As in-depth questions, Figure 7 As shown, questions such as "Did you meet him smoothly?" and "Is ○○ okay?" can be asked.

[0118] Similarly, when the destination of the user 80 is the classroom and the accuracy of the user 80's behavior (movement) estimated by the behavior estimation unit 65 is calculated to be high, the question selection unit 66 selects questions related to movement that are more in-depth than the questions when the accuracy is medium. Figure 7 As shown in FIG, questions such as “Are you sweating all over?” and “Are you feeling good?” can be given. Similarly, when the accuracy of the action (exercise) of the user 80 estimated by the action estimation unit 65 is calculated to be high, the question selection unit 66 selects questions related to exercise that are more in-depth than the questions when the accuracy is medium. As in-depth questions, Figure 7 As shown, questions such as "Have you become proficient?" and "Have you understood?" can be asked.

[0119] In this way, by calculating the accuracy of the behavior of the user 80 estimated by the behavior estimation unit 65 and determining questions based on the calculated accuracy, the conversation with the user 80 can be activated.

[0120] Furthermore, when calculating the accuracy of the estimated action, the accuracy calculation unit 69 may also refer to the schedule data of the user 80. By referring to the schedule data, the accuracy of the calculation can be improved. In the above description, the accuracy of the estimated action (meal) is low when the action history of the user 80 stored in the action history database 53b is zero. In this case, if the schedule data includes information such as "dining at a restaurant," the action estimation unit 65 can also calculate the accuracy of the estimated action (meal) as medium or high by referring to the schedule data.

[0121] Next, refer to Figure 8 An example of the structure of questions stored in the question database 53c will be described.

[0122] like Figure 8 As shown, questions are stored in the question database 53c in a manner classified into multiple levels (first level to third level). The question selection unit 66 selects questions belonging to a certain level based on the accuracy of the estimated action calculated by the accuracy calculation unit 69. Specifically, when the accuracy of the estimated action is less than the first specified value, the question selection unit 66 selects questions included in the first level. Figure 8 As shown, the questions included in the first level are questions asking about the actions of the user 80 themselves.

[0123] When the accuracy of the estimated action is greater than or equal to the first predetermined value and less than or equal to the second predetermined value (first predetermined value < second predetermined value), the question selection unit 66 selects questions included in the second level. Figure 8 As shown, the questions included in the second level are questions used to confirm the estimated actions.

[0124] When the accuracy of the estimated action is greater than the second predetermined value, the question selection unit 66 selects questions included in the third level. Figure 8 As shown, the questions included in the third level are questions that ask the user to answer their thoughts or comments on the actions before boarding the bus. The third level can also include questions that ask the user to answer their thoughts or comments on POIs. As questions that ask the user to answer their thoughts or comments on the actions before boarding the bus, if the action before boarding the bus is to eat, then Figure 8 As shown in , questions such as "Is it delicious?" and "Do you want to eat here again?" can be given. As questions for users to answer their thoughts or comments on POI, if the attribute of POI is a restaurant, then Figure 8 As shown, questions such as "Is the store clean?" can be asked.

[0125] As Figure 8 The first and second predetermined values shown in FIG. 6 may also use the probabilities calculated by the accuracy calculation unit 69. For example, the first predetermined value may be 30% and the second predetermined value may be 70%. The first level corresponds to Figure 7 The second level is equivalent to Figure 7 The third level is equivalent to Figure 7 The estimation accuracy is high.

[0126] As the user 80 moves from the first level to the third level, that is, as the accuracy of the estimated action increases, the questions become more in-depth. The actions of the user 80 at the POI are recorded in the action history database 53b, and the action history is accumulated. As the accumulation of action history increases, the accuracy of the estimated action calculated by the accuracy calculation unit 69 increases, and the questions become more in-depth. In this way, the more the accumulation of action history increases, the more active the conversation with the user 80 can be. In addition, Figure 8 In the , questions are categorized into three levels, but this is not limited to them. Questions can also be further refined, such as at the fourth and fifth levels. In other words, the levels can be set based on the degree of abstraction or specificity of the question, with questions assigned to the levels being more abstract at higher levels and more specific at lower levels.

[0127] Next, refer to Figure 9The relationship between the action history records stored in the action history database 53 b and the accuracy of the estimated action calculated by the accuracy calculation unit 69 will be described.

[0128] As described above, the fewer the action history records stored in the action history database 53b, the lower the accuracy of the estimated action. On the other hand, the more the action history records stored in the action history database 53b, the higher the accuracy of the estimated action.

[0129] Similarly, the lower the probability of an action history stored in action history database 53b, the lower the accuracy of the estimated action. On the other hand, the higher the probability of an action history stored in action history database 53b, the higher the accuracy of the estimated action. In the above description, if 7 of 10 action history records indicate dining and 3 indicate discussion, the probability that user 80's action at the restaurant was dining is 70%. However, this 70% probability refers to the probability of the action history records.

[0130] Furthermore, the accuracy calculation unit 69 may calculate the accuracy of the estimated action without referring to the action history records stored in the action history database 53b. If the attribute of the POI is a shopping mall, as described above, there are many actions that the user 80 can perform, making it difficult to estimate the action performed by the user 80. Consequently, the accuracy of the estimated action decreases. On the other hand, if the attribute of the POI is a ramen restaurant, the action that the user 80 may perform is determined to be dining, so the accuracy of the estimated action increases. If the attribute of the POI is a supermarket, the action that the user 80 may perform is roughly determined to be shopping, so the accuracy of the estimated action is moderate. In this way, the accuracy calculation unit 69 can also calculate the accuracy of the estimated action using only the attributes of the POI, without referring to the action history records stored in the action history database 53b.

[0131] Next, refer to Figure 10 The following is a flowchart of an operation example of the information processing device 101 according to the modification example. Among them, the processing of steps S201 to S207 and S219 to S225 is the same as Figure 5 The processes of steps S101 to S107 and S111 to S117 are similar, and therefore their description is omitted.

[0132] In step S209 , the accuracy calculation unit 69 calculates the accuracy indicating the accuracy (probability) of the behavior of the user 80 estimated by the behavior estimation unit 65 .

[0133] The process proceeds to step S211, and the question selection unit 66 selects questions belonging to a certain level based on the accuracy of the estimated action calculated by the accuracy calculation unit 69 (see Figure 8Specifically, if the accuracy of the estimated behavior is less than the first specified value, the question selection unit 66 selects a question that inquires about the behavior of the user 80 itself (step S213). Alternatively, if the accuracy of the estimated behavior is greater than the first specified value and less than the second specified value, the question selection unit 66 selects a question for confirming the estimated behavior (step S215). Alternatively, if the accuracy of the estimated behavior is greater than the second specified value, the question selection unit 66 selects a question that asks the user to answer questions about their thoughts or evaluations of their actions before boarding the vehicle or questions that ask the user to answer questions about their thoughts or evaluations of the POI (step S217).

[0134] (Effect)

[0135] As described above, according to the information processing device 101 according to the modification example, the following operational effects can be obtained.

[0136] The controller 60 calculates the accuracy of the estimated behavior based on the behavior history or POI attributes stored in the behavior history database 53b and determines questions based on the calculated accuracy of the estimated behavior. This can avoid asking questions that are off-topic.

[0137] Questions are categorized into multiple levels (refer to Figure 8 The controller 60 determines questions belonging to a certain level as output data based on the accuracy of the estimated action. By determining questions based on the accuracy of the estimated action, the conversation with the user 80 can be activated. Furthermore, the controller 60 may determine the abstractness of the questions based on the accuracy of the estimated action and determine the determined questions as output data.

[0138] The output data includes questions for the user to answer about pre-boarding activities, questions for the user to answer about thoughts or comments about pre-boarding activities, and questions for the user to answer about thoughts or comments about POIs. By using various questions, the conversation with the user 80 can be activated.

[0139] The controller 60 classifies the user 80's answer as either a positive or negative answer, and records the classification result as classification data in the action history database 53b, in association with the POI. By classifying the user 80's answer as a positive or negative answer, the user 80 can be asked a more in-depth question the next time.

[0140] Controller 60 can also output classification data associated with POIs based on the location information of vehicle 40. Suppose user 80 responds affirmatively to a meal at ramen shop 90, such as "It's delicious." In this case, when vehicle 40 passes near ramen shop 90, controller 60 can output a voice message such as "The ramen is delicious!" This can invigorate the conversation with user 80.

[0141] Each of the functions described in the above embodiments can be implemented by one or more processing circuits. Processing circuits include programmed processing devices, such as processing devices containing electrical circuits. Processing circuits also include devices such as application-specific integrated circuits (ASICs) and circuit components configured to perform the described functions.

[0142] While the embodiments of the present invention have been described above, it should not be understood that the description and drawings constituting part of this disclosure limit the present invention. Various alternative embodiments, examples, and application techniques will be readily apparent to those skilled in the art based on this disclosure.

[0143] exist Figure 3 In the example shown, the vehicle 40 is parked in the parking lot of the ramen shop 90, so the location information of the vehicle 40 is regarded as the location information of the POI. However, it is not limited to the destination where there is a parking lot. In the case where there is no parking lot at the destination, for example, consider the case where the user 80 Figure 11 As shown, the user 80 parks in a coin-operated parking lot near the destination and walks from the coin-operated parking lot to the destination. In this case, the location information of the vehicle 40 does not match the location information of the destination (the location information of the POI). Therefore, even if the location information of the vehicle 40 is stored in association with the user 80's pre-boarding behavior, it will not be useful data.

[0144] Therefore, the information processing device 100 may also store the location information of the terminal held by the user 80 (hereinafter referred to as the user terminal) in association with the user 80's pre-boarding actions in the action history database 53b. Specifically, when the user 80 arrives at the destination, the information processing device 100 obtains the location information of the user terminal through communication. In addition, as a prerequisite, it is assumed that the information processing device 100 and the user terminal can communicate. As an example, the information processing device 100 includes a receiver for receiving data transmitted from the user terminal. The information processing device 100 can determine whether the user 80 has arrived at the destination by referring to the information of the guidance application installed in the user terminal.

[0145] The information processing device 100 estimates the user 80's behavior at the destination based on the location information acquired from the user terminal. The information processing device 100 then asks questions related to the estimated behavior. Furthermore, the information processing device 100 associates the location information acquired from the user terminal with the user 80's pre-boarding behavior and stores it in the behavior history database 53b. This allows the user 80's behavior at the POI to be accurately recorded, even if the vehicle 40's parking location is far from the destination (the location of the POI).

[0146] Description of Reference Numerals

[0147] 40: Vehicle; 50: Sensor group; 51: GPS receiver; 52: Microphone; 53: Storage device; 53a: Map database; 53b: Action history database; 53c: Question database; 54: Speaker; 60: Controller; 61: User determination unit; 62: Destination arrival judgment unit; 63: Sound analysis unit; 64: Action history update unit; 65: Action estimation unit; 66: Question selection unit; 67: Occupancy detection unit; 68: Question output unit; 69: Accuracy calculation unit; 100, 101: Information processing device.

Claims

1. An information processing device, characterized in that have: Controller; a vehicle boarding detection device for detecting a user's boarding of a vehicle; a storage device for recording the user's action data; an output device for outputting question data to be answered by the user; as well as an input device that receives input from the user, wherein, when the controller detects that the user has boarded the vehicle according to the signal obtained from the boarding detection device, the controller determines the attributes of the point of interest based on the position information of the vehicle, The controller estimates the user's behavior before boarding the vehicle based on the attributes of the point of interest, The controller calculates the accuracy of the estimated behavior of the user before boarding the vehicle based on the attribute of the point of interest. The controller outputs, from the output device, output data including at least a question regarding the user's behavior before boarding the vehicle based on the calculated accuracy of the estimated behavior, The controller obtains the user's answer to the question as input data via the input device, The controller stores the input data in the storage device in association with the position information of the vehicle or the point of interest.

2. The information processing device according to claim 1, wherein The points of interest are any one of points of interest located around the location information of the vehicle, a destination set in a navigation device of the vehicle, points of interest included in data related to the user's planned action obtained through communication, and points of interest included in data related to the user's pre-boarding action history obtained through communication.

3. The information processing device according to claim 1 or 2, characterized in that The controller determines the question according to an attribute of the point of interest determined based on the position information of the vehicle.

4. The information processing device according to claim 1 or 2, characterized in that The storage device records the attributes of the point of interest and the action categories into which the actions that the user may perform at the point of interest are classified in association with each other. The controller determines the question based on the type of action of the user associated with the attribute of the point of interest.

5. The information processing device according to claim 1 or 2, characterized in that The controller acquires data related to the user's planned action through communication, The question is determined by referring to data related to the user's planned action.

6. The information processing device according to claim 1 or 2, characterized in that further comprising a receiver for receiving data transmitted from a terminal held by the user, The controller determines the question based on the position information of the terminal before the boarding transmitted from the terminal.

7. The information processing device according to claim 1, wherein The controller determines the abstractness of the question according to the accuracy of the estimated action, and determines the determined question as output data.

8. The information processing device according to claim 1 or 2, characterized in that The output data includes questions for the user to answer regarding the actions before boarding the vehicle, questions for the user to answer regarding thoughts or evaluations regarding the actions before boarding the vehicle, or questions for the user to answer regarding thoughts or evaluations regarding the points of interest.

9. The information processing device according to claim 1 or 2, characterized in that The controller classifies the user's answer into an affirmative answer or a negative answer, and records the classification result as classification data in the storage device in association with the point of interest.

10. The information processing device according to claim 1 or 2, characterized in that The input data is data obtained by converting the user's voice, or data generated by receiving an operation from the user.

11. The information processing device according to claim 1 or 2, characterized in that The controller detects the user getting off the vehicle after the vehicle arrives at the destination set by the user, and then outputs the question when the user's getting on the vehicle is detected.

12. The information processing device according to claim 9, wherein The controller outputs the classification data associated with the point of interest according to position information of the vehicle.

13. An information processing method using an information processing device for recording the behavior of a user riding a vehicle, characterized in that: include: determining an attribute of a point of interest based on position information of the vehicle when the user's boarding of the vehicle is detected based on a signal obtained from a boarding detection device; estimating the user's pre-ride behavior based on the attributes of the point of interest; calculating the accuracy of the estimated behavior of the user before boarding the vehicle based on the attributes of the point of interest; outputting output data including at least a question regarding the user's pre-ride behavior based on the calculated accuracy of the estimated behavior; Obtaining the user's answer to the question as input data; as well as The input data is stored in a storage device in association with the position information of the vehicle or the point of interest.

Images