WO2024189877A1

WO2024189877A1 - Information processing device, information processing method, information processing program, and storage medium

Info

Publication number: WO2024189877A1
Application number: PCT/JP2023/010266
Authority: WO
Inventors: 秦松崎
Original assignee: パイオニア株式会社
Priority date: 2023-03-16
Filing date: 2023-03-16
Publication date: 2024-09-19

Abstract

One objective of the present invention is to provide an information processing device, an information processing method, an information processing program, and a storage medium that make it possible to provide many pieces of submission information submitted from general users in an easy-to-handle form or to generate easy-to-handle information on the basis of submission information.　The present invention is characterized by comprising: a submission group acquisition unit that acquires one or more submission groups into which a plurality of pieces of submission information each including submission content submitted from a mobile body and a submission location that is the location at which the submission was made are divided by submission content type on the basis of the submission location; and a knowledge generation unit that, for each submission group, generates knowledge that is based on the submission location and the submission content type, wherein the knowledge generation unit generates the knowledge in the form of points, areas, or lines on the basis of the submission content type.

Description

Information processing device, information processing method, information processing program, and storage medium

The present invention relates to an information processing device, an information processing method, an information processing program, and a storage medium, and, for example, to an information processing device, an information processing method, an information processing program, and a storage medium that processes information posted on a network.

There is a known technology that associates location information, which indicates the location of the poster when the post was posted on a network, with the post and stores it in a server. For example, Patent Document 1 discloses a post providing system that corrects location information based on the content of the post and extracts and provides posts that match search criteria.

JP 2013-50917 A

One of the issues with the above-mentioned systems for providing posted messages is that even if individual posted messages that match the search criteria are provided, it can be difficult to provide useful or reliable information to users.

The present invention has been made in consideration of the above points, and one of its objectives is to provide an information processing device, an information processing method, an information processing program, and a storage medium that can provide a large amount of posted information from general users in an easy-to-handle form, or that can generate easy-to-handle information based on the posted information.

The invention described in claim 1 is an information processing device that includes a post group acquisition unit that acquires one or more post groups in which multiple pieces of post information, including post content posted from a mobile body and the post location where the post was made, are divided based on the post location for each type of post content, and a knowledge generation unit that generates knowledge for each post group based on the post location and the type of post content, and the knowledge generation unit generates the knowledge in the form of a point, area, or line based on the type of post content.

The invention described in claim 10 is an information processing method executed by an information processing device, comprising: a post group acquisition step of acquiring one or more post groups in which a plurality of pieces of post information, including post content posted from a mobile body and a post location where the post was made, are divided based on the post location for each type of the post content; and a knowledge generation step of generating knowledge for each post group based on the post location and the type of the post content, wherein in the knowledge generation step, the knowledge is generated in the form of a point, an area, or a line based on the type of the post content.

The invention described in claim 11 is an information processing program executed by an information processing device having a computer, which causes the computer to execute a post group acquisition step of acquiring one or more post groups in which a plurality of pieces of post information, including post content posted from a mobile body and a post location where the post was made, are divided for each type of post content based on the post location, and a knowledge generation step of generating knowledge for each post group based on the post location and the type of post content, and in the knowledge generation step, generating the knowledge in the form of a point, area, or line based on the type of post content.

The invention described in claim 12 is a computer-readable storage medium that stores an information processing program for causing an information processing device equipped with a computer to execute a post group acquisition step of acquiring one or more post groups in which multiple pieces of post information, including post content posted from a mobile body and the post location where the post was made, are divided based on the post location for each type of post content, and a knowledge generation step of generating knowledge for each post group based on the post location and the type of post content, and generating the knowledge in the form of a point, area, or line based on the type of post content in the knowledge generation step.

1 is a schematic diagram illustrating an overview of an information processing system according to an embodiment of the present invention. 1 is a diagram showing a configuration of a front seat portion of an automobile according to a first embodiment; 1 is a block diagram showing an example of a configuration of an in-vehicle device according to a first embodiment; FIG. 4 is a diagram illustrating an example of posted information according to the first embodiment. FIG. 2 is a block diagram showing an example of a configuration of a server device according to the first embodiment. FIG. 11 is an explanatory diagram of a distance threshold in clustering according to the first embodiment. FIG. 4 is a diagram illustrating an example of cluster information according to the first embodiment; FIG. 11 is an explanatory diagram of correction of a distance threshold in the clustering according to the first embodiment. FIG. 11 is an explanatory diagram of correction of a distance threshold in the clustering according to the first embodiment. 5 is a flowchart illustrating an example of a routine executed by the server device according to the first embodiment. FIG. 13 is a diagram showing an example of a knowledge mode setting for each content type according to the second embodiment; FIG. 11 is a diagram illustrating an example of knowledge information according to the second embodiment. FIG. 11 is a diagram illustrating an example of knowledge information according to the second embodiment. FIG. 11 is a diagram illustrating an example of knowledge information according to the second embodiment. FIG. 11 is a diagram showing an example in which point knowledge according to the second embodiment is displayed superimposed on a map. FIG. 11 is a diagram showing an example in which range knowledge according to the second embodiment is displayed superimposed on a map. FIG. 11 is a diagram showing an example in which line knowledge according to the second embodiment is displayed superimposed on a map. 5 is a flowchart illustrating an example of a routine executed by the server device according to the embodiment. 5 is a flowchart illustrating an example of a routine executed by the server device according to the embodiment.

The following describes in detail an embodiment of the present invention. In the following description and accompanying drawings, the same reference symbols are used for substantially the same or equivalent parts.

The configuration of an information processing system 100 including an information processing device according to the first embodiment will be described with reference to the attached drawings.

FIG. 1 shows an overview of the configuration of the information processing system 100. As shown in FIG. 1, the information processing system 100 includes a server 10 as an information processing device, and an in-vehicle device 30.

The in-vehicle device 30 is a terminal device equipped with a communication function, such as a car navigation device or a drive recorder, mounted on a moving body. Note that FIG. 1 shows a case in which the in-vehicle device 30 is mounted on an automobile M, which is an example of a moving body.

The server 10 and the in-vehicle device 30 can transmit and receive data to and from each other via the network NW using communication protocols such as TCP/IP and UDP/IP. The network NW can be constructed, for example, by a mobile communication network, wireless communication such as Wi-Fi (registered trademark), and Internet communication including wired communication.

The information processing system 100 is an information processing system that collects and performs statistical processing on posted information posted from multiple terminal devices, including the in-vehicle device 30. When each of the multiple terminal devices receives a post input from a user, it generates posted information including the posted content and a posted position indicating the position of the terminal device at the time of posting, and transmits the information to the server 10. The server 10 performs clustering on the posted information posted from each of the terminal devices using the distance between the posted positions, and generates one or more clusters (post groups) of posted information.

In this embodiment, clustering is performed for each piece of posted information indicating the content of the post. Therefore, in this embodiment, each piece of posted information used for clustering is assigned a type of posted content (hereinafter also referred to as a content type) that classifies the posted content into one of a number of predetermined types, or the type of posted content is assigned by the server 10. The server 10 performs the above-mentioned clustering for each type of posted content. For example, the posted information includes information indicating a type of content, such as "tourist spot" for a place where monuments suitable for tourism are installed, "rain" for the weather, "patrol car" for a sighting of a police car on patrol, and "stop line" and "road construction" for roads.

Furthermore, the server 10 can generate knowledge based on the type of post content and the posting position for each cluster generated by the above clustering. The knowledge includes, for example, for the cluster on which the knowledge is based, information on the position indicated by a point, line, or area based on the type of post content of posts included in the cluster and the posting position of posts included in the cluster. The position indicated by this point, line, or area is also referred to below as the position of the knowledge.

In this embodiment, the server 10 can transmit knowledge that can be displayed superimposed on a map image to a terminal device such as an in-vehicle device 30. The in-vehicle device 30 can display the knowledge on a display screen inside the vehicle, and notify the user of information based on the knowledge by screen display or voice. For example, the in-vehicle device 30 can detect when it is approaching the location of the knowledge or when it has entered the area of the knowledge, and notify the user of information according to the content of the knowledge.

The posted information collected by the server 10 includes posted information posted from a moving object. In this specification, posted information posted from a moving object includes posted information posted from a terminal device mounted on the moving object such as the in-vehicle device 30, as well as posted information posted from a portable terminal device such as a smartphone, tablet, or wearable device used by a user riding in a moving object such as an automobile, motorcycle, or bicycle. It may also include posted information posted from a terminal device used by a pedestrian.

In this embodiment, an in-vehicle device 30 is taken as an example of a terminal device that transmits posted information to the server 10 or receives knowledge information generated from the posted information. In addition, in this embodiment, an example is described in which the in-vehicle device 30 is a car navigation device. In addition, in this embodiment, an example is described in which the in-vehicle device 30 is a terminal device of a so-called cloud-type car navigation device that receives, from a user, a destination to which the user wishes to be guided, transmits the destination to the server 10, and the server 10 generates a route to the destination.

FIG. 2 is a perspective view showing the vicinity of the front seat of an automobile M equipped with an in-vehicle device 30. As an example of an installation, FIG. 2 shows a case where the in-vehicle device 30 is installed in the dashboard DB of the front seat of the automobile M.

The GPS receiver 11 is a device that receives signals (GPS signals) from GPS (Global Positioning System) satellites. The GPS receiver 11 is arranged, for example, on a dashboard DB. The GPS receiver 11 may be arranged anywhere as long as it can receive GPS signals. The GPS receiver 11 is capable of transmitting the received GPS signals to the in-vehicle device 30. The in-vehicle device 30 obtains current position information of the automobile M using the GPS signals.

The touch panel 13 is, for example, a touch panel monitor that combines a display, such as a liquid crystal display capable of displaying images, with a touch pad. The touch panel 13 is disposed, for example, in the center console of the dashboard DB. The touch panel 13 may be disposed in a location that is visible to the driver and within the driver's reach. For example, the touch panel 13 may be attached to the dashboard DB.

The touch panel 13 is capable of displaying a screen based on the control of the in-vehicle device 30. The touch panel 13 is also capable of transmitting a signal representing an input operation to the touch panel 13 received from a user to the in-vehicle device 30. For example, the touch panel 13 may display car navigation guidance. Furthermore, it may be possible to perform operations related to the car navigation function, such as setting a destination, via the touch panel 13.

The touch panel 13 also displays screens such as a screen for accepting input of the message to be posted and the type of post content, and a screen for displaying map knowledge in which knowledge sent from the server 10 is superimposed on a map.

The speaker 15 is provided, for example, on the interior side of the A-pillar AP. The speaker 15 is capable of emitting sounds such as music and voice based on the control of the in-vehicle device 30. For example, the speaker 15 emits the voice of the navigation system.

In addition, in this embodiment, for example, a voice notification regarding map knowledge transmitted from the server 10 is emitted from the speaker 15. For example, a voice explaining the knowledge displayed on the touch panel or a voice notifying that the automobile M is approaching the knowledge is emitted from the speaker 15.

The microphone 17 is a microphone device that picks up sounds inside the vehicle, and is located, for example, on the dashboard DB. The microphone 17 may be located anywhere, such as the rear-view mirror RM or the steering wheel, as long as it can pick up sounds inside the vehicle.

In this embodiment, for example, posts by an occupant of the automobile M, who is the user of the in-vehicle device 30, are made by voice via the microphone 17. For example, the voice picked up by the microphone 17 may be converted into text information and used as the content of the posted information, or a voice file may be included in the posted information.

The exterior camera 18 is an imaging device that captures images of the surroundings of the automobile M. For example, the exterior camera 18 is attached to the inside of the windshield FG near the rearview mirror RM so that the imaging direction is toward the front. For example, the exterior camera 18 can capture images of the front of the automobile M through the windshield. The exterior camera 18 may be disposed on the dashboard DB, for example. The automobile M may be equipped with a camera that can capture images of the sides or rear of the automobile M.

The in-vehicle camera 19 is an imaging device that captures images of the interior of the automobile M. In this embodiment, the in-vehicle camera 19 is configured integrally with the exterior camera 18. For example, the in-vehicle camera 19 is configured to be able to capture images of the interior of the automobile M, including the driver's seat, passenger seat, and rear seats of the automobile M. For example, an image captured by the exterior camera 18 or the in-vehicle camera 19 may be attached to the posted information.

FIG. 3 is a block diagram showing the configuration of the in-vehicle device 30. For example, the in-vehicle device 30 is a device in which a storage unit 23, a control unit 25, and a communication unit 27 work together via a system bus.

The automobile M is also equipped with an acceleration sensor 21. The acceleration sensor 21 is capable of measuring the acceleration of the automobile M and outputting a signal indicating the measured acceleration. The acceleration sensor 21 is a sensor capable of detecting the acceleration in the direction of travel of the automobile M when viewed from above the automobile M, that is, the forward/backward direction. The acceleration sensor can also detect, for example, the acceleration in the lateral direction (width direction) perpendicular to the direction of travel of the automobile M.

The storage unit 23 is a storage device configured, for example, by a hard disk drive, a solid state drive (SSD), a flash memory, etc. The storage unit 23 stores various programs executed in the in-vehicle device 30, such as an operating system and software for the terminal.

The various programs may be obtained, for example, from another server device or the like via a network, or may be recorded on a recording medium and read via various drive devices. In other words, the various programs stored in the storage unit 23 can be transmitted via a network, and can also be recorded on a computer-readable recording medium and transferred.

The memory unit 23 also stores map information including road maps. The map information is used, for example, to display guidance and knowledge for car navigation.

The control unit 25 is composed of a CPU (Central Processing Unit), ROM (Read Only Memory), RAM (Random Access Memory), etc., and functions as a computer. The control unit 25 realizes various functions by the CPU reading and executing various programs stored in the ROM and the storage unit 23. In this embodiment, the control unit 25 performs functions such as generating post information in response to input from the user and transmitting it to the server 10, and car navigation functions.

The control unit 25 is connected to be able to communicate with various devices provided in the automobile M, namely, the GPS receiver 11, the touch panel 13, the speaker 15, the microphone 17, the exterior camera 18, the interior camera 19, and the acceleration sensor 21. The control unit 25 acquires data from the various devices provided in the automobile M. The control unit 25 also supplies data to the various devices provided in the automobile M.

Specifically, the control unit 25 sequentially acquires GPS signals from the GPS receiver 11. The control unit 25 also sequentially acquires signals indicating acceleration measured by the acceleration sensor 21. The control unit 25 acquires, for example, information indicating the current position information and moving direction of the automobile M based on the GPS signals and signals from the acceleration sensor. The control unit 25 can also acquire the moving trajectory of the automobile M based on the current position of the automobile M.

The control unit 25 acquires signals representing input operations performed on the touch panel 13. The control unit 25 also supplies image data to be displayed on the touch panel 13. The control unit 25 also supplies audio data to the speaker 15. The control unit 25 also acquires audio in the automobile M picked up by the microphone 17. The control unit 25 acquires images captured by the exterior camera 18 and the interior camera 19.

The control unit 25 generates post information in response to an input operation performed by a user of the in-vehicle device 30. For example, when the user starts a posting application on the in-vehicle device 30 and inputs the content to be posted by voice via the microphone 17, the control unit 25 generates post information including the voice data of the content to be posted or text data converted from the voice data, and a posting location, which is information indicating the location of the in-vehicle device 30 at the time of posting.

In addition, when text input of the post content is made via the touch panel 13 instead of voice input of the post content via the microphone 17, the control unit 25 generates post information including the text data of the post content and the posting position. In addition, the control unit 25 accepts the selection of the type of post content via the microphone 17 or the touch panel 13.

The communication unit 27 is a communication device that transmits and receives data to and from external devices in accordance with instructions from the control unit 25. The communication unit 27 is, for example, a NIC (Network Interface Card) for connecting to the network NW. The communication unit 27 is connected to the above-mentioned network NW, and transmits and receives various data to and from the server 10.

The control unit 25 transmits the generated post information to the server 10 via the communication unit 27. Also, for example, the control unit 25 can transmit information including a destination input by a user of the in-vehicle device 30 to the server 10 via the communication unit 27, and receive route information or navigation information to the destination from the server 10.

The control unit 25 receives knowledge information generated by the server 10, for example, via the communication unit 27, and displays the received knowledge on the touch panel 13 by superimposing it on the map image.

FIG. 4 is a diagram showing an example of posted information generated by the control unit 25 and transmitted to the server 10 as PD1. For example, the posted information is information in which the posting time and ID are associated with the posting location indicating the position of the in-vehicle device 30 at the time of posting, i.e., the position of the automobile M in this embodiment, and the posted content. In other words, the posted information is information that includes the posted content posted from the terminal device and the posting location, which is the location where the post was made.

As described above, the posted content may be, for example, text data input by the poster via a touch panel, text data converted from speech uttered by the poster, or audio data of the speech. In other words, the posted content may include information generated by speech uttered by the poster.

In addition, in this embodiment, the post information includes information indicating the type of post content, which classifies the post content into one of the specified types. The type of post content is a type into which post content related to the location indicated by the posting position is classified. The type of post content is selected by the user, for example, when inputting the post content. Also, for example, if the post content is not assigned a type or the type assigned to the post content is inappropriate, the control unit 25 may obtain the type of post content based on keywords included in the voice data or text data converted from the voice when the post content is input by voice, and generate the post information.

The posted information may also include information indicating the moving state of the automobile M, such as information indicating whether the automobile M was moving or stopped at the time of posting, information indicating the moving direction and speed, and information indicating the moving trajectory. In this case, the posted information can be said to include the posting content posted from the moving body, the posting location where the posting was made, and moving state information indicating the moving state of the moving body at the time the posting was made.

In the example shown in FIG. 4, the posted information is categorized as "rain" and includes information indicating that the information was posted while moving (movement type), and movement status information indicating the speed and direction. The movement type may be determined from the speed. The posted information may include, for example, an image captured by the exterior camera 18 or the interior camera 19 at the time of posting.

FIG. 5 is a block diagram showing the configuration of the server 10. For example, the server 10 is a device in which a large-capacity storage device 31, a communication unit 33, and a control unit 35 work together via a system bus.

As described above, the server 10 has a function to perform clustering of the posted information posted from each terminal device into one or more post groups according to the type of posted content, and generate one or more clusters (post groups). The server 10 also has a function to generate knowledge for each cluster.

The server 10 also has the function of receiving current location information of the automobile M from the in-vehicle device 30 and information on a destination set by a user who is an occupant of the automobile M, and generating a route to the destination based on the current location information and the destination information.

The large-capacity storage device 31 is composed of, for example, a hard disk device and an SSD (Solid State Drive), and stores various programs such as an operating system and software for the server 10. The various programs may be obtained, for example, from other server devices, etc., via a network, or may be recorded on a recording medium and read via various drive devices. In other words, the various programs stored in the large-capacity storage device 31 can be transmitted via a network, and can also be recorded on a computer-readable recording medium and transferred.

The large-capacity storage device 31 stores, for example, an information processing program that allows the server 10 to execute clustering. The large-capacity storage device 31 also stores, for example, an information processing program that allows the server 10 to execute knowledge generation for each cluster.

The large-capacity storage device 45 also includes various databases used for clustering and knowledge generation. The map information database (map information DB in the figure) 31A is a map information database in which map information including road maps is stored. For example, the map information in the map information DB is used by the server 10 to generate map knowledge, which is image information in which knowledge is superimposed on a map. The map information in the map information DB is also used by the server 10 when generating routes for car navigation.

The post DB31B stores the posted information posted from each of the terminal devices. As described above, the posted information posted from each of the terminal devices includes posted information posted from a mobile object.

The cluster DB 31C stores information indicating the clusters generated as a result of the clustering performed by the server 10. Specifically, the cluster DB 31C stores information in which the posted information ID of the posted information included in each cluster is associated with a cluster ID that identifies each cluster.

The knowledge DB 31D stores knowledge information indicating the knowledge generated by the server 10 for each cluster. The knowledge information includes the type of posted content of the posted information included in the original cluster, and the location information of the knowledge indicating the location of the original cluster as a whole.

In addition, each of the various databases described above may be stored in an external server separate from server 10, and server 10 may read the necessary information from the external server.

The communication unit 33 is connected to the network NW described above, and transmits and receives various data to and from terminal devices such as the in-vehicle device 30.

The control unit 35 is composed of a CPU (Central Processing Unit), ROM (Read Only Memory), RAM (Random Access Memory), etc., and functions as a computer. In the control unit 35, the CPU realizes various functions by reading and executing various programs stored in the ROM and the large-capacity storage device 31. In this embodiment, the control unit 35 performs functions such as a function of performing clustering on a large number of posted information, a function of generating knowledge for each cluster generated by clustering, and a car navigation function.

When the control unit 35 receives posted information from a terminal device via the communication unit, it stores the received posted information in the posting DB.

The control unit 35 has, as a functional unit, a clustering unit 37 as a post group generation unit that acquires posted information from the post DB and performs clustering on the acquired posted information. The control unit 35 also has, as a functional unit, a knowledge generation unit 39 that generates knowledge for each cluster generated by the clustering unit 37. The clustering performed by the clustering unit 37 in this embodiment will be described in detail below.

[Clustering]
First, the clustering unit 37 acquires posted information to be used for clustering from the posted information DB. For example, the clustering unit 37 acquires posted information whose posting positions are within a predetermined area from the posted information DB, and performs clustering for each predetermined area.

As described above, among the posted information posted from terminal devices, posted information from a mobile object may include movement status information.

The clustering unit 37 functions as a posting information acquisition unit that executes a step of acquiring posting information posted from a moving object as posting information to be used for clustering, and acquiring posting information including the posting content posted from the moving object, the posting location where the posting was made, and movement state information indicating the movement state of the moving object at the time the posting was made (posting information acquisition step).

When the clustering unit 37 acquires the posted information, it performs clustering for each type of posted content. In this clustering, the clustering unit 37 groups posted information having the same type of posted content and in which the distance between the posted positions is equal to or less than a threshold into the same cluster, thereby dividing one or more pieces of posted information of the same type into one or more clusters.

The clustering unit 37 generates clusters of posted information by, for example, Euclidean clustering using the Euclidean distance between posting positions. Specifically, a cluster including one or more pieces of posted information is generated by assigning the same cluster ID to posted information whose posting positions are within a threshold distance.

In this embodiment, the distance threshold is determined at least according to the type of post content. For example, in the case of a type of post content related to weather, posts tend to be made over a wide range about the same subject, such as rain in the same area, and a relatively long distance threshold (e.g., several hundred meters) is set. For example, in the case of a type related to road markings such as stop lines or tourist spots where monuments are installed, it is expected that posts about the same subject will be made within a relatively narrow range, and a relatively short distance threshold (e.g., a dozen meters) is set.

By setting the distance threshold in this way, posted information about the same object even within the same category can be grouped into the same cluster, while posted information about different objects (for example, different monuments at the same tourist spot) can be separated into different clusters. This makes it possible to use clusters as data that is easier to handle compared to simply displaying posted information, for example, by displaying posted information by cluster. Furthermore, clusters can be used to generate knowledge based on clusters, making it possible to provide users with more useful information.

FIG. 6A is a diagram that shows, together with a map, a schematic diagram of how multiple pieces of posted information P1 to P7, which were posted with points in a certain area (area MP1) as the posting location, are clustered by type of posted content. In FIG. 6A, icons representing posted information P1 to P7 are shown at the positions indicated by the posting locations included in each piece of posted information. In FIG. 6A, "Stop" (stop line), "Patrol" (police car), and "Rain" (rain) indicate the types of posted content. In the example shown in FIG. 6A, posted information P1 to P7 are divided into five clusters with cluster IDs (cluster numbers) 1 to 5.

In FIG. 6A, in clustering of posted information P1 and P2, whose posted content type (content type) is "Stop," the distance threshold is set to TH1. Because the distance between the posting location of posted information P1 and the posting location of adjacent posted information P2 of the same type is longer than the distance threshold TH1, posted information P1 and posted information P2 are divided into separate clusters and assigned different cluster numbers "1, 2."

In addition, in FIG. 6A, the distance threshold is set to TH2 in the clustering of posted information P3 to P5, whose content type is "Patrol." As shown in FIG. 6A, the distance between the posting location of posted information P3 and the posting location of adjacent posted information P4 of the same type is less than or equal to the distance threshold TH2, so posted information P3 and posted information P4 are in the same cluster and are assigned the same cluster number "3."

Because the posting location of posted information P5 is farther away from the posting locations of posted information P3 and posted information P4 than the distance threshold TH2, posted information P5 is classified into a different cluster from posted information P3 and P4 and is assigned the cluster number "4." In other words, posted information P5 is clustered into a cluster that includes only posted information P5 because the distance to the nearest posting location of the same content type exceeds the distance threshold TH2.

In addition, in FIG. 6A, the distance threshold is set to TH3 in the clustering of posted information P6 to P7 of the content type "Rain." As shown in FIG. 6A, the distance between the posting positions of posted information P6 and posted information P7 of the same type is less than or equal to the distance threshold TH3, so they are assigned the same cluster number "5" as they are in the same cluster.

As described above, the clustering distance threshold TH1 for the type "pause" where the posted content has a local target can be set relatively short, and the clustering distance threshold TH3 for the type "rain" where the posted content has a wide range of target can be set relatively long. In the example shown in FIG. 6A, TH1 < TH2 < TH3. In this way, by performing clustering using a distance threshold determined according to the type of posted content, posted information about the same target is more likely to be included in the same cluster, and posted information about different targets is more likely to be separated into different clusters.

When the control unit 35 generates a cluster, it generates cluster information that associates a cluster ID, which is an identifier that identifies each cluster, with the post ID of the post information included in the cluster indicated by the cluster ID, and stores the cluster information in the cluster DB 31C.

FIG. 6B is a diagram showing cluster information for clusters with cluster IDs 1 to 5 in FIG. 6A as CD1, as an example of cluster information. As shown in FIG. 6B, each cluster ID is associated with the post ID of the post information included in the cluster. For example, the cluster information is used when knowledge is generated for each cluster.

In this embodiment, in order to make it easier to calculate the distance between posting locations in the Euclidean clustering process, each posting location is converted from a geographic coordinate system to a planar coordinate system before clustering. For example, after clustering and knowledge generation, when the knowledge is displayed superimposed on a map, the knowledge location is converted back to a geographic coordinate system.

In addition, in FIG. 6A, for the sake of explanation, the posting position and clusters are displayed together with a map, but in this embodiment, clustering is processed independently of map information. In this embodiment, clustering is performed without correcting the position to match a map, for example, using map matching. In this way, it is possible to support various types of maps. Specifically, as one example, it is possible to prevent a problem in which information that is incompatible with a terminal having a map different from the specific map information is generated by correcting the position using specific map information.

For example, if clustering or knowledge generation is performed while making corrections to suit a specific map, the processing will be tailored to the type of map used for clustering or knowledge generation, so it is expected that the position will shift when the data is displayed superimposed on another type of map. In this embodiment, position corrections to suit a specific map are not performed during clustering or knowledge generation, so the above-mentioned position shifts can be prevented. In addition to such position shifts, there are roads whose positions have changed depending on whether the map is new or old, so by not making corrections based on a specific map, it becomes possible to generate knowledge regardless of whether the map is new or old.

[Distance Threshold Correction]
When performing clustering, the clustering unit 37 can correct the distance threshold for the posted information posted while moving, based on the movement state information included in the posted information.

FIG. 7 is a diagram showing an example of setting the distance threshold in the clustering of posted information P3 and posted information P4 described in FIG. 6A, in which posted information P3 is posted while moving. In FIG. 7, posted information P3 is assumed to have been posted from a moving object moving in the direction of the arrow. In this case, posted information P3 includes information indicating that it was posted while moving.

In this way, when the posted information P3 indicates that it was posted while moving, the clustering unit 37 sets the distance threshold for the posted information posted while moving to a distance threshold TH4 that is longer than the distance threshold TH2 based on the content type.

As shown in FIG. 7, the distance between the posting locations of posted information P3 and posted information P4 exceeds TH2 but is within the distance threshold TH4. Therefore, by clustering using the corrected distance threshold TH4, posted information P3 and posted information P4 are included in the same cluster.

For example, it is believed that there is a similar time lag between the time the poster sees an object, such as a police car, and the time the poster posts about that object, whether the information is posted while moving or while stopped. This means that the distance from the location where the object is located to the posting location of information posted while moving tends to be greater than the distance from the location where the object is located to the posting location of information posted while stopped. Specifically, because posts made while moving are made while moving away from the object, the posting location of a post made while moving is often farther from the object that is the subject of the post than the posting location of a post made while stopped.

Therefore, for posts posted while moving, the distance between the posting locations for the same subject may be longer than the distance between the posting locations for posts posted while stationary. In such cases, the distance between the posting locations for the same subject may exceed the distance threshold, and posts about the same subject may be separated into different clusters.

As described above, by adjusting the distance threshold for clustering for posted information posted while moving, it is possible to make it easier for posted information about the same subject to be included in the same cluster, for example.

Furthermore, when the posted information includes speed information indicating the speed of the moving object at the time the post was made, the control unit 35 as a post group generation unit may correct the distance threshold to a length according to the speed information. For example, the faster the speed indicated by the speed information is, the longer the distance threshold may be corrected, and when the speed exceeds a predetermined speed (e.g., 40 km/h), the distance threshold may be corrected to a longer value.

As described above, the posted information may include, as the posted content, information generated by the voice spoken by the poster. The control unit 35, which serves as a post group generating unit, may correct the distance threshold based on the time taken to speak when posting the posted information and on speed information indicating the speed of the moving object at the time the post was made.

For example, if a poster posts by speaking while moving, there will be a difference between the location where the target object is located and the posting location by the time it took to speak. Therefore, for example, the faster the speed, the greater the distance between the location where the target object is located and the posting location due to the time it took to speak.

As a result, posted information about the same subject may be divided into different clusters. As described above, by correcting the distance threshold based on the time and speed information required for utterance, posted information about the same subject can be more likely to be included in the same cluster.

Figure 8 is a diagram showing an example of distance threshold correction for posted information posted while moving. In Figure 8, posted information P8 and P9 are posted information of a tourist spot type ("Tour" in the figure) that targets monument 40. Posted information P8 is posted from mobile body M1 equipped with in-vehicle device 30 while moving, and the arrow in Figure 8 indicates the traveling direction of mobile body M1. Posted information P9 is posted from, for example, a portable terminal while stopped.

The posted information P8 includes, as movement state information, movement direction information that indicates the direction of movement of the moving object at the time of posting. In addition, the distance threshold for the content type "Tour" is TH5, and the range within the distance threshold TH5 from the posting location of the posted information P8 is shown as AR5 in FIG. 8.

The clustering unit 37 may generate clusters based on the movement direction information included in the posted information P8 by setting a distance threshold in the direction opposite to the movement direction from the posting position of the posted information P8 to a longer distance than the distance threshold in the movement direction indicated by the movement direction information.

Specifically, for example, the distance threshold in the direction indicated by the movement direction information, i.e., the traveling direction of the moving body M1, is set to TH5, and for the direction opposite to the traveling direction, a distance threshold TH6 longer than the distance threshold TH5 is set, and clustering is performed. For example, the clustering unit 37 sets an elliptical range AR6 as a range in which the distance from the posting position of the posted information P8 in the traveling direction of the moving body M1 is within the distance threshold TH5 and the distance in the direction opposite to the traveling direction is within the distance threshold TH6, and posts whose posting position is within AR6 are placed in the same cluster as the posted information P8.

In the example shown in FIG. 8, the posting location of posted information P9 is not within the distance threshold TH5 (within range AR5) from the posting location of posted information P8. However, the posting location of posted information P9 is within range AR6, which takes into account the distance threshold TH6 in the direction opposite to the direction of travel, so posted information P9 is considered to be in the same cluster as posted information P8 and is assigned the same cluster ID "7".

For example, when posting information while moving, it is possible that the object is in the opposite direction to the direction of travel, and that there are also more posts about the same object on the opposite side of the direction of travel. As described above, by correcting the distance threshold for information posted while moving, taking into account the direction of travel, it is possible to make it easier for posted information about the same object to be included in the same cluster.

The range AR6 in which the distance from the posting position of the posted information P8 in the direction of travel of the moving body M1 is within the distance threshold TH5 and the distance in the direction opposite the direction of travel is within the distance threshold TH6 is not limited to the elliptical range described above. The range AR6 may be of any shape as long as it extends longer in the direction opposite the direction of travel than the direction of travel of the moving body M1. The clustering unit may set the range AR6 to, for example, a rectangle, a rounded rectangle, an egg shape, etc.

In addition, instead of correcting the distance threshold for the posted information posted while moving as described above, the clustering unit 37 may correct the posting position of the posted information posted while moving based on the moving speed or moving direction, and then perform clustering.

The clustering unit 37 may generate clusters by performing clustering for each piece of posted information whose moving direction information indicates the same direction, based on the moving direction information included in the posted information. For example, whether the moving directions are the same or not may be determined, for example, by setting a reference direction and judging the angle with respect to the reference direction. For example, even if the directions are not exactly the same, a predetermined tolerance may be set and posted information within the tolerance may be clustered together. For example, when map information is used, whether the moving directions are the same or not may be determined by distinguishing between uphill and downhill directions on a road.

For example, when it comes to content types related to roads, such as stop lines and road construction, there is a high possibility that posted information about the same subject will move in the same direction, and by clustering posts in the same direction, posted information about the same subject will be more likely to be included in the same cluster.

As described above, the clustering unit 37 sets a distance threshold between posting locations and groups posts that are equal to or less than the distance threshold into the same post group, thereby performing clustering to separate multiple posted information into one or more post groups (clusters).

For posted information posted from a moving object and including movement state information, the distance threshold is set for each type of posted content, and is set according to information regarding the movement state of the moving object, such as whether the posted information was posted while moving, the speed, the direction of movement, etc.

In other words, the clustering unit 37 functions as a post group generation unit that executes a step of dividing a plurality of pieces of posted information into one or more post groups (post group generation step) using the distance between the posting positions for each type of posted information.

The clustering unit 37, which serves as a post group generating unit, classifies post information whose distance between posting locations is equal to or less than a distance threshold determined according to the type of post content and movement state information into the same post group when generating the post group.

The control unit 35 can also receive, for example, posted information posted from a terminal device, including the posted content and the posted location, and is not limited to information posted from a mobile object. In other words, the clustering unit 37 as a posted information acquisition unit can function as a posted information acquisition unit that acquires posted information including the posted content posted from a terminal device and the posted location, which is the location where the post was made.

In addition, in clustering, which classifies multiple pieces of posted information posted from a terminal device into one or more groups (clusters) of posts by grouping posts that are equal to or smaller than a distance threshold into the same group, the distance threshold is set for each type of post content.

In other words, the clustering unit 37, which functions as a post group generation unit, generates post groups by grouping posts whose distance between posting positions is equal to or less than a distance threshold determined according to the type of post content.

Clustering Control Routine
9 is a flowchart showing a cluster generation routine RT1, which is an example of a control routine executed by the control unit 35 of the server 10. For example, when the server 10 is powered on, the control unit 35 causes the clustering unit 37 to repeatedly execute the cluster generation routine RT1.

When the cluster generation routine RT1 starts, the clustering unit 37 determines whether new posted information is stored in the posting DB 31B (step S101). In step S101, for example, the clustering unit 37 determines whether a predetermined number or more pieces of new posted information that have not been clustered have been received.

If it is determined in step S101 that no new post information is stored in the post DB 31B (step S101: NO), the clustering unit 37 ends the cluster generation routine RT1 and starts a new cluster generation routine RT1.

In step S101, if it is determined that new posted information is stored in the posting DB 31B (step S101: YES), the clustering unit 37 acquires posted information to be used for clustering from the posting DB 31B (step S102). In step S102, the clustering unit 37 acquires posted information posted from a moving object, for example, as shown in FIG. 4A.

The posted information includes, for example, the posted content, the posted location which is the location of the moving object when the post was made, and movement state information which indicates the movement state of the moving object when the post was made. In step S102, for example, all of the posted information stored in the post DB 31B that has not been clustered is read out. In step S102, the clustering unit 37 functions as a posted information acquisition unit.

After executing step S102, the clustering unit 37 converts the posting location of each piece of posted information acquired in step S102 from the geographic coordinate system to the planar coordinate system (step S103).

After executing step S103, the clustering unit 37 determines a distance threshold for each type of posted content based on the posted information, and performs correction according to the movement state information as necessary (step S104). In step S104, for example, the clustering unit 37 sets the distance threshold from the posting position of posted information posted while moving to a longer distance than the distance threshold from the posting position of posted information posted while stopped.

After step S104 is executed, the clustering unit 37 executes clustering for each type of post content using the distance threshold set in step S104 (step S105). In step S105, post information whose posting positions are separated by the distance equal to or less than the threshold is classified as being in the same cluster, and one or more post information of the same type is divided into one or more clusters.

In steps S104 and S105, the clustering unit 37 functions as a post group generation unit that divides a plurality of pieces of posted information into one or more post groups using the distance between the posting positions for each type of posted content. In generating post groups, the clustering unit 37 as the post group generation unit groups posted information whose distance between the posting positions is equal to or less than a distance threshold determined according to the type of posted content and the movement state information into the same post group.

After executing step S105, the clustering unit 37 generates cluster information including the correspondence between each cluster and the posted information contained in each cluster, and stores the cluster information in the cluster DB 31C (step S106). After executing step S106, the clustering unit 37 ends the cluster generation routine RT1, and starts a new cluster generation routine RT1.

In step S106 of this routine, for example, in addition to the cluster information, the posted information included in the cluster (for example, the posted information whose posting position has been subjected to coordinate conversion in step S103) may be stored in the cluster DB 31C. Furthermore, the cluster information and the posted information may be transmitted to the knowledge generation unit 39. In this case, the knowledge generation unit 39 generates knowledge for each cluster generated in this routine.

As described above, the information processing device of this embodiment has a post information acquisition unit that acquires post information including post content posted from a moving object, a post location where the post was made, and movement state information indicating the movement state of the moving object when the post was made, and a post group generation unit that divides the multiple pieces of post information into one or multiple post groups (clusters) using the distance between the post locations for each type of post content.

In generating post groups (clustering), the post group generating unit classifies post information whose distance between post locations is equal to or less than a distance threshold determined according to the type of post content and movement state information into the same post group.

With the above configuration, for example, an appropriate distance threshold can be set for each type of post content. As a result, even if the type of post content is the same, posts that are likely to be about different subjects can be separated into different post groups, and posts that are likely to be about the same subject can be included in the same post group.

Furthermore, in the case of posted information while moving, there is a concern that there may be a large discrepancy between the posted location and the location where the target object or event of interest is occurring. In this embodiment, by setting a distance threshold according to the movement state information as described above, even if posted information is posted while moving, posts that are highly likely to be about the same target can be included in the same post group.

Therefore, for example, the clusters generated by the information processing device of this embodiment can be treated as a single set of data for each cluster, and can be used for various purposes. For example, for multiple posted information, it is possible to generate information that is easier to understand than simply displaying individual pieces of information, such as by displaying the posting locations for each cluster on a map. Also, as described above, knowledge can be generated for each cluster by the knowledge generation unit 38.

Therefore, according to this embodiment, it is possible to provide an information processing device, an information processing method, an information processing program, and a storage medium that can provide a large amount of posted information posted by general users in an easy-to-handle form, or generate easy-to-handle information based on the posted information.

The configuration and functions of an information processing system 100 including a server 10 as an information processing device of Example 2 will be described with reference to Figures 10 to 16. The information processing system 100 of Example 2 is configured similarly to the information processing system 100 of Example 1. As described above, the control unit 35 of the server 10 has a knowledge generation unit 39 as a functional unit. The knowledge generation unit 39 is capable of generating knowledge for each cluster.

As described above, knowledge is, for example, information that includes the characteristics of the cluster that is the source of the knowledge, such as the location of the cluster as a whole and the type of post content.

The clusters used to generate knowledge in Example 2 include one or more clusters (post groups) in which multiple pieces of post information posted from a mobile body, including the post content and the post location where the post was made, are divided by type of post content based on the post location.

In Example 2, the knowledge generation unit 39 generates knowledge for each cluster based on the posting position and the type of posting content of the posting information included in the cluster. In generating the knowledge, the knowledge generation unit 39 generates the knowledge in the form of a point, an area, or a line based on the type of posting content for each cluster.

[Knowledge Generation]
When generating knowledge, the knowledge generating unit 39 first acquires a cluster to be used for knowledge generation. For example, the knowledge generating unit 39 refers to the cluster information stored in the cluster DB 31C, identifies the posting ID of the posting information included in the cluster, and acquires the posting information corresponding to the posting ID from the posting DB 31B. Alternatively, the knowledge generating unit 39 receives the cluster information and the posting information from the clustering unit 37.

The knowledge generation unit 39 functions as a post group acquisition unit that executes a step of acquiring one or more clusters (post groups) in which multiple pieces of post information posted from a mobile body, the post contents and the post positions where the posts were made, are divided based on the post positions for each type of post contents (post group acquisition step).

FIG. 10 shows an example of a knowledge mode setting table, TB2, used when generating knowledge. TB2 is stored, for example, in the mass storage device 31. In TB2, recommended knowledge modes are associated with each type of post content (content type).

As described above, when generating knowledge, the knowledge generation unit 39 uses clusters that are clustered by type, so the types of post content included in the clusters are consistent. After identifying the content type for each cluster, the knowledge generation unit 39 determines the form of knowledge to be generated, for example, by referring to TB2.

In TB2, when the content types are "tourist attractions," "stop lines," and "traffic accidents," point knowledge is recommended. For these content types, even if a cluster contains multiple posted information, they tend to be concentrated in a relatively small area, and it is appropriate to represent the locations of multiple posts as a single point, so the point knowledge mode is recommended.

In addition, in TB2, when the content types are "Police car," "Road construction," "Frozen road," and "Tree-lined street," line-like knowledge is recommended. For these content types, when a cluster contains multiple posted information, it is assumed that the posted information tends to be posted along roads, for example, and it is appropriate to represent the multiple posting positions with lines, so a line-like knowledge mode is recommended.

In addition, in TB2, when the content types are "rain," "wind," "theft," and "event venue," knowledge of the range aspect is recommended. For these content types, when multiple posted information is included in a cluster, it is assumed that there is a tendency for them to be distributed over a wide area, and it is appropriate to express them as a range that includes multiple posting locations, so the knowledge of the range aspect is recommended.

Note that the knowledge types for each content type shown in TB2 are just examples, and knowledge types that are considered appropriate can be set appropriately depending on the characteristics and distribution trends of the content type.

FIGS. 11A to 11C are diagrams showing examples of knowledge generated by the knowledge generation unit 39 as knowledge information ND1 to ND3. The knowledge information ND1 to ND3 is information in which a knowledge ID that identifies each piece of knowledge is associated with the state of the knowledge, the location of the knowledge, and the type of posted content. As shown in the knowledge information ND1 to ND3, the knowledge may also include an expiration date for the knowledge.

Expiration dates are set for each type of post content. For example, for content types "Police cars," "Frozen roads," and "Weather," the situation is likely to change in a relatively short period of time, so a relatively short expiration date of, for example, a few hours to a day is set. For content types "Road construction" and "Theft," the situation is likely to change in a certain period of time, so an expiration date of, for example, a few days to a few weeks is set.

For example, tourism-related content types such as "tourist spots" and "tree-lined streets" are thought to remain unchanged for a relatively long period of time, so an expiration date of, for example, six months to several years may be set. For example, in the case of a content type that is also tourism-related but whose status changes in a short period of time, such as an "event venue" for a festival, a relatively short expiration date of, for example, a few hours to a few days may be set.

Knowledge information ND1 shows an example of point knowledge in the form of a point for the content type "pause line." The point knowledge is generated, for example, as a point located at the center of gravity of the posting positions of one or more pieces of posted information included in a cluster. The position of the point knowledge is represented by the position of the center of gravity.

Also, as shown in ND1, the knowledge may be assigned a movement direction. For example, if the movement state information of the posted information included in a cluster includes a movement direction and the movement directions are consistent, the consistent movement direction may be assigned to the knowledge. Also, as described above, even if clustering is performed for each movement direction, the movement direction of the posted information included in the cluster is common, so the common movement direction may be assigned to the knowledge.

For example, if the knowledge is of a content type that only pertains to one lane of a road, such as a stop line or road construction, the knowledge will be unnecessary information for a user of a terminal device traveling in the opposite direction to the direction of movement of the knowledge. Therefore, it is preferable not to transmit the knowledge to such a terminal device.

By assigning a direction of movement to knowledge, it is possible to control the sending of knowledge only to terminal devices moving in the same direction as the direction of movement of the knowledge, and not to users moving in the opposite direction. This makes it possible to provide a better user experience by appropriately sending the necessary information to the user.

Knowledge information ND2 shows an example of range knowledge in the form of a range for the content type "rain". The range knowledge is generated, for example, as a range that surrounds all of the posting positions of the posted information included in the cluster, for example, a range of any shape, such as a rectangle, a circle, an ellipse, etc. The position of the range knowledge is represented by the coordinates of two points on a diagonal line in the case of a rectangular range, for example.

Knowledge information ND3 shows an example of line knowledge of the line pattern for the content type "frozen road surface". The line knowledge is generated, for example, as a line connecting the posting positions of multiple posted information included in a cluster, or an interpolation straight line or curve of multiple posting positions. The position of the line knowledge is represented, for example, by the coordinates of multiple points indicating the line knowledge. For example, the multiple points may be connected by a straight line, and smoothing may be performed on the terminal device that displays the line connecting the multiple points by straight lines.

The knowledge generation unit 39 stores the generated knowledge information in the knowledge DB 31D. For example, the knowledge generation unit 39 generates map knowledge that can be displayed by being superimposed on a map. For example, when knowledge is generated by the knowledge generation unit 39, the control unit 35 transmits the knowledge information of the knowledge or the map knowledge to the terminal device.

Figures 12 to 14 show an example of a display screen in which knowledge of each aspect of points, areas, and lines is transmitted to an in-vehicle device 30, which is an example of a terminal device, and is displayed superimposed on a map on the touch panel 13.

FIG. 12 shows an example in which point knowledge N1 to N3 are displayed on the touch panel 13. For the sake of explanation, FIG. 12 also shows posted information P11 to P16 contained in the original clusters 1 to 3 of each of the point knowledge N1 to N3, but when the point knowledge N1 to N3 are actually displayed, the posted information P11 to P16 is not displayed.

As in point knowledge N1 in FIG. 12, point knowledge is generated as one piece of knowledge even when the original cluster contains only one piece of posted information. Also, as described above, point knowledge is generated as a point located at the center of gravity of the posting positions of one or more pieces of posted information contained in the cluster. For example, point knowledge N2 is generated in the form of a point located at the midpoint between two posting positions, and point knowledge N3 is generated in the form of a point located at the center of gravity of three posting positions.

FIG. 13 shows an example in which range knowledge N4 to N6 are displayed on the touch panel 13. For example, the range knowledge is generated as a range that encloses the posting positions of one or more pieces of posted information included in the original cluster, extending far enough outward from the extreme posting positions.

FIG. 14 shows an example in which line knowledge N7-N8 and point knowledge N9 are displayed on the touch panel 13. As described above, the line knowledge is generated as a line connecting the posting positions of multiple pieces of posted information included in a cluster, or as an interpolation straight line or curve between multiple posting positions. For example, the line knowledge N7 is generated in the form of a line connecting two posting positions. The line knowledge N8 is generated as an interpolation curve between three posting positions.

For example, as shown in FIG. 14, even if a content type is one for which line knowledge is recommended, the clustering unit 37 generates point knowledge if the cluster contains only one piece of posted information.

[Correction process]
As in the first embodiment, the posted information posted from the moving object includes information indicating the moving direction and speed of the moving object at the time the posting was made, or moving status information indicating the moving trajectory of the moving object.

When generating knowledge, the knowledge generation unit 39 may perform correction processing based on the movement state information of the moving object at the time the posted information was posted. For example, the knowledge generation unit 39 performs pre-generation correction processing to generate knowledge by correcting the position of each piece of posted information included in a group of posts (cluster) based on the movement state information. Also, for example, the knowledge generation unit 39 performs post-generation correction processing to correct the position of the generated knowledge based on the movement state information. Thereby, for example, it is possible to reduce the influence of the movement state on the distance from the position where the poster viewed the object of the post to the posting position, and to improve the accuracy of the final position of the knowledge.

For example, when knowledge is generated, the posted information included in a cluster is considered to include information posted while moving in various directions at various speeds, as well as information posted while stopped. For example, if a cluster contains a large number of posted information, it is considered that the influence of posted information posted while moving on the distance from the position where the poster viewed the posted object to the posted position will be small, for example, due to cancellation. In other words, the smaller the number of posted information, the more likely it is that the posting position of posted information posted while moving will affect the position of the knowledge.

Then, when the number of posted information pieces included in each cluster does not reach a predetermined number, the knowledge generation unit 39 may perform the above-mentioned correction process, i.e., pre-generation correction process or post-generation correction process.

The knowledge generation unit 39 may also generate knowledge according to the distribution state of the posting positions of the posting information included in the cluster, in addition to the type of posting content. For example, for a cluster of a content type for which point knowledge is recommended, if the posting positions are distributed linearly, line knowledge may be generated in addition to point knowledge. Also, for a cluster of a content type for which point knowledge is recommended, if the posting positions are distributed over a wide range, range knowledge may be generated in addition to point knowledge.

[Transmission of knowledge to terminal device]
For example, when the knowledge generating unit 39 generates knowledge, the knowledge generating unit 39 stores knowledge information including the location and content type of the knowledge in a knowledge DB. The knowledge generating unit 39 can also generate image information in which the knowledge is superimposed on a map as map knowledge.

When the knowledge is generated, the control unit 35 of the server 10 transmits the knowledge to the in-vehicle device 30. For example, the control unit 35 transmits map knowledge to the in-vehicle device 30. Also, for example, the control unit 35 may transmit only knowledge information that does not include map information to the in-vehicle device 30. In that case, for example, the knowledge superimposed on the map may be displayed on the in-vehicle device 30 based on the map information held by the in-vehicle device 30. For example, the map knowledge or knowledge information is transmitted to an in-vehicle device 30 located in an area close to the location of the knowledge.

The control unit 35 can, for example, notify the contents of the knowledge to a terminal device approaching the location of the knowledge. For example, the server 10 acquires current location information of the in-vehicle device 30 while moving, and judges whether the terminal device T has approached the location of the knowledge using criteria according to the type of knowledge. For example, in the case of line knowledge, the shortest distance from the current location of the in-vehicle device 30 to the line knowledge can be used as the criterion, and it can be determined that the in-vehicle device 30 has approached the location of the knowledge when the shortest distance falls within a predetermined distance.

Or, for example, in the case of line knowledge, when the in-vehicle device 30 is moving in a direction along the line knowledge, if the shortest distance from the current position of the in-vehicle device 30 to the line knowledge falls within a predetermined distance, the terminal device T may determine that it is approaching the position of the line knowledge. Also, when the in-vehicle device 30 is moving in a direction that intersects with the line knowledge, even if the shortest distance from the current position of the in-vehicle device 30 to the line knowledge falls within a predetermined distance, the terminal device T may determine that it is not approaching the position of the line knowledge, and may not notify the in-vehicle device 30 of the contents of the knowledge.

For example, if the route of the in-vehicle device 30 only intersects with line knowledge, the in-vehicle device 30 is considered to be less affected by the content of the post about the line knowledge, and the line knowledge is considered to be unnecessary information for the user of the in-vehicle device 30. By determining the state of the line knowledge as described above, it is possible to control so as not to send unnecessary knowledge to the user.

For example, if the route along which the in-vehicle device 30 plans to travel intersects with the line knowledge multiple times, the in-vehicle device 30 may determine that it is approaching the position of the line knowledge. Furthermore, if the line knowledge does not follow a road, the contents of the knowledge may be notified even if the route of the in-vehicle device 30 does not follow the line knowledge.

In the case of point knowledge, the determination is based on the distance from the current position of the in-vehicle device 30 to the position of the point knowledge. In the case of range knowledge, the determination may be based on, for example, the shortest distance from the current position of the in-vehicle device 30 to the range knowledge, or the distance to the center or center of gravity of the range knowledge.

The control unit 35 determines whether a moving user is approaching the location of the knowledge using criteria that correspond to the state of the knowledge, and when it determines that the user is approaching the location of the knowledge, it functions as a notification unit that notifies the terminal device used by the user.

[Knowledge Update]
When a new post is made, the control unit 35 adds the new posted information, performs clustering again, and generates knowledge for each of the newly generated clusters, thereby updating the knowledge. For example, when there is knowledge that has the same content type and is located close to the posting location of the new posted information, the clustering unit 37 acquires the posted information included in the original cluster of the knowledge from the posting DB 31B based on the cluster information, and adds the new posted information to the acquired information to perform reclustering. The knowledge generating unit 39 generates knowledge for each cluster generated by the reclustering.

[Knowledge reliability]
Furthermore, the knowledge generating unit 39 may set the reliability of the knowledge. For example, the knowledge generating unit 39 may set the reliability of the knowledge based on the expiration date of the knowledge described above. For example, the knowledge generating unit 39 may set the reliability of the knowledge based on the number of posted information included in the cluster used to generate the knowledge. For example, in calculating the reliability by weighted average, the reliability may be calculated by weighting the number of posted information and the remaining time until the expiration date.

[Knowledge Generation Control Routine]
15 is a flowchart showing a knowledge generation routine RT2, which is an example of a control routine executed by the control unit 35 of the server 10. For example, when the server 10 is powered on, the control unit 35 causes the knowledge generation unit 39 to repeatedly execute the knowledge generation routine RT2.

When the knowledge generation unit 39 starts the knowledge generation routine RT2, it determines whether or not there is a new cluster that has not been converted into knowledge (step S201). In step S201, for example, it is determined whether or not new cluster information has been saved in the cluster DB 31C.

If it is determined that there are no clusters that have not been converted into knowledge (step S101: NO), the knowledge generation unit 39 ends the knowledge generation routine RT2 and starts a new knowledge generation routine RT2.

If it is determined that there is a cluster that has not been converted into knowledge (step S101: YES), the knowledge generation unit 39 acquires the post information included in that cluster from the post DB 31B (step S202). In step S202, for example, the knowledge generation unit 39 acquires the post information corresponding to the post ID associated with the cluster ID based on the cluster information in the cluster DB 31C.

In addition, in steps S201 and S202, the knowledge generation unit 39 may obtain the clusters by receiving the cluster information and the posting information from the clustering unit 37.

In step S202, the knowledge generation unit 39 functions as a post group acquisition unit that acquires one or more post groups in which multiple pieces of post information, including the post content posted from a mobile body and the post location where the post was made, are divided based on the post location for each type of post content.

After executing step S202, the knowledge generation unit 39 identifies the type of post content for each cluster (step S203).

After executing step S203, the knowledge generation unit 39 determines the knowledge aspect for each cluster based on the type of post content identified in step S203 (step S204). In step S204, for example, the knowledge generation unit 39 refers to a setting table such as the one illustrated in FIG. 10, and identifies the knowledge aspect recommended for each content type.

In step S204, for example, the knowledge mode may be determined according to the type of post content as well as the distribution state of the posting positions of one or more pieces of post information included in the cluster.

After executing step S204, the knowledge generation unit 39 converts the posting location of each piece of posted information acquired in step S202 from the geographic coordinate system to the planar coordinate system (step S205).

Note that, in step S202, if the knowledge generation unit 39 receives posted information in which the posting position has been converted into a planar coordinate system from the clustering unit, it is not necessary to execute step S205.

After executing step S205, the knowledge generation unit 39 generates knowledge for each cluster (step S206). In step S206, for example, the knowledge generation unit 39 converts the posting position of the posting information included in the cluster into position information of the type determined in step S204, and sets it as the position of the knowledge.

In step S206, the knowledge generation unit 39 generates knowledge in the form of a point, an area, or a line based on the type of the posted content. In step S206, knowledge information including information indicating the position of the knowledge and the type of the posted content is generated for each cluster.

After executing step S206, the knowledge generation unit 39 converts the position of the knowledge from the planar coordinate system to the geographic coordinate system. Through steps S206 and S207, knowledge information such as that shown in Figures 11A to 11C is generated and stored in the knowledge DB.

In steps S203 to S207, the knowledge generation unit 39 functions as a knowledge generation unit that generates knowledge for each post group (cluster) based on the posting location and the type of posting content.

After execution of step S207, the knowledge generation unit 39 generates map knowledge, which is image data for displaying the knowledge generated in step S206 and subjected to coordinate conversion in step S207, superimposed on a map (step S208). In step S208, for example, the knowledge generation unit 39 uses the map information in the map information DB 31A to generate a display image of the knowledge, such as those illustrated in Figures 12 to 14.

After executing step S208, the knowledge generation unit 39 stores the map knowledge in the knowledge DB 31D (step S209), and then ends the knowledge generation routine RT2.

[Update Control Routine]
16 is a flowchart showing a reclustering routine RT3 which is an example of a control routine executed by the control unit 35 of the server 10. For example, when knowledge is generated, the control unit 35 causes the clustering unit 37 to repeatedly execute the reclustering routine RT3.

When the clustering unit 37 starts the reclustering routine RT3, it determines whether or not a new post has been received by the server 10 (step S301). If it is determined in step S301 that there is no new post (step S301: NO), the clustering unit 37 ends the reclustering routine RT3 and starts a new reclustering routine RT3.

In step S301, when it is determined that there is a new post (step S301: YES), the clustering unit 37 searches the knowledge DB 31D (step S302) and determines whether there is knowledge that has the same content type as the new post and is located close to the posting position of the new post (within a specified distance) (step S303). For example, the specified distance is set to be longer than the distance threshold for clustering.

In step S303, if it is determined that there is knowledge of the same type and in a nearby location as the new post (step S303: YES), the clustering unit 37 refers to the cluster DB 31C and obtains the post ID of the post information included in the original cluster of the knowledge (step S304).

After executing step S304, the clustering unit 37 acquires the post information corresponding to the post ID acquired in step S303 from the post DB 31B (step S305).

After executing step S305, the clustering unit 37 executes clustering using the posted information acquired in step S305 and the new information acquired in step S301 (step S306). In step S306, for example, steps S103 to S106 of the cluster generation routine RT1 are executed to generate clusters.

After executing step S306, the clustering unit 37 ends the reclustering routine RT3 and starts a new reclustering routine RT3.

For example, when the reclustering routine RT3 is executed, the knowledge generation unit 39 executes the knowledge generation routine RT2, and knowledge is generated for each cluster generated by the reclustering. In this way, the knowledge is updated.

As described above, the information processing device of Example 2 has a post group acquisition unit that acquires one or more post groups in which a plurality of pieces of post information, including post content posted from a mobile body and the post location where the post was made, are divided based on the post location for each type of post content, and a knowledge generation unit that generates knowledge for each post group based on the post location and the type of post content. The knowledge generation unit generates knowledge in the form of a point, area, or line based on the type of post content.

According to this embodiment, for example, knowledge can be generated in a form appropriate for the type of post content. For example, by presenting the generated knowledge to the user, it is possible to provide the user with more useful information than if the posted information were simply displayed.

The configurations, routines, etc. of the server 10 and the in-vehicle device 30 in the above-described embodiment are merely examples and can be appropriately selected or modified depending on the application, etc.

In the above embodiment, the in-vehicle device 30 has been described as an example of a terminal device that generates posted information and transmits it to the server 10, but terminal devices that are not mounted on a moving body can also generate posted information and transmit it to the server 10. For example, a portable terminal device has sensors such as a GPS sensor and an acceleration sensor, and can obtain movement status information and include it in the posted information. Note that posted information posted from a terminal device that does not move, such as a fixed PC, does not need to include movement status information.

In the above embodiment, the in-vehicle device 30 is an in-vehicle navigation device, but the in-vehicle device 30 does not need to have a navigation function. In that case, for example, the in-vehicle device 30 may be configured to be able to include information on the current position and speed of the automobile M in the posted information and transmit it to the server 10.

For example, the in-vehicle device 30 may be configured by integrating a terminal device having a similar configuration to the in-vehicle device 30 with the exterior camera 18, the interior camera 19, and the touch panel 13. Specifically, for example, the in-vehicle device 30 may be a terminal device such as a camera-equipped smartphone, tablet, or PC equipped with an application that performs the same functions as the in-vehicle device 30. In this case, the in-vehicle device 30 may be mounted on the dashboard DB, for example, by a cradle, etc., so that the built-in camera can capture images of the area in front of the automobile M through the windshield of the automobile M.

Furthermore, the in-vehicle device 30 may be configured not to display a screen to be presented to the driver of the automobile M. For example, the in-vehicle device 30 may have a configuration similar to that of a drive recorder, and may be a device integrated with the exterior camera 18 and the interior camera 19. In this case, the in-vehicle device 30 may not perform the various display outputs as described above.

10 Server 30 Vehicle-mounted device 11 GPS receiver 13 Touch panel 15 Speaker 17 Microphone 18 Vehicle exterior camera 19 Vehicle interior camera
21 Acceleration sensor 31 Large-capacity storage device 31A Map information DB
31B Post DB
31C Cluster DB
31D Knowledge DB
33 Communication unit 35 Control unit 37 Clustering unit 38 Knowledge generation unit

Claims

a post group acquisition unit that acquires one or more post groups in which a plurality of pieces of post information including post content posted from a mobile object and a post location where the post was made are classified based on the post location for each type of the post content;
a knowledge generating unit configured to generate knowledge for each of the post groups based on the post position and the type of the post content,
The information processing apparatus, wherein the knowledge generating unit generates the knowledge in the form of any one of a point, an area, and a line based on a type of the posted content.
the posted information includes movement state information that is information indicating a moving direction and a speed of the moving object when the post was made or information indicating a moving trajectory of the moving object when the post was made,
The information processing device according to claim 1, characterized in that the knowledge generation unit generates the knowledge by correcting the position of each piece of posted information included in the post group based on the movement status information, or performs a correction process to correct the position of the generated knowledge.
The information processing device according to claim 2, characterized in that the knowledge generation unit performs the correction process when the number of pieces of posted information included in each of the post groups does not reach a predetermined number.
The information processing device according to claim 1, characterized in that the knowledge generation unit generates knowledge according to the distribution state of the posting positions of one or more of the posted information included in the post group in addition to the type of the posted content.
The information processing device according to claim 1, further comprising a notification unit that determines whether a moving user is approaching the location of the knowledge using a determination criterion according to the state of the knowledge, and notifies a terminal device used by the user when it is determined that the user is approaching the location of the knowledge.
The post group acquisition unit acquires a new post group generated for a plurality of pieces of posted information including new posted information posted after the knowledge is generated;
The information processing apparatus according to claim 1 , wherein the knowledge generating unit generates new knowledge for the new post group based on the posting position and the type of the posting content.
The information processing device according to claim 1, characterized in that the knowledge generation unit sets an expiration date for each piece of knowledge based on the type of the posted content.
The information processing device according to claim 7, characterized in that the knowledge generation unit sets the reliability of the knowledge based on the expiration date.
The information processing device according to claim 1, characterized in that the knowledge generation unit sets the reliability of the knowledge based on the number of pieces of posted information included in the post group used to generate the knowledge.
An information processing method executed by an information processing device,
a post group acquisition step of acquiring one or more post groups in which a plurality of pieces of post information including post contents posted from a mobile object and a post location where the post was made are classified based on the post location for each type of the post contents;
a knowledge generating step of generating knowledge for each of the post groups based on the posting positions and the types of the posting contents,
The information processing method, wherein in the knowledge generating step, the knowledge is generated in the form of any one of a point, an area, and a line based on the type of the posted content.
An information processing program executed by an information processing device having a computer, the information processing program comprising:
a post group acquisition step of acquiring one or more post groups in which a plurality of pieces of post information including post contents posted from a mobile object and a post location where the post was made are classified based on the post location for each type of the post contents;
a knowledge generating step of generating knowledge for each of the posts based on the posting position and the type of the posting content;
Run the command,
An information processing program for causing the step of generating the knowledge in the form of any one of a point, an area, or a line based on the type of the posted content in the knowledge generating step.
An information processing device including a computer,
a post group acquisition step of acquiring one or more post groups in which a plurality of pieces of post information including post contents posted from a mobile object and a post location where the post was made are classified based on the post location for each type of the post contents;
a knowledge generating step of generating knowledge for each of the posts based on the posting position and the type of the posting content;
Run the command,
A computer-readable storage medium that stores an information processing program for executing, in the knowledge generation step, generating the knowledge in the form of any one of a point, an area, or a line based on the type of the posted content.