Information processing device

Abstract

The present invention keeps map data in a useful state.　By using a natural language processing model, this information processing device confirms the topic of a topic confirmation point, which is at least a part of points registered in map data, and edits the map data on the basis of the topic of the topic confirmation point.

Description

Information processing apparatus

[0001] The present invention relates to an information processing apparatus.

[0002] Map data used in navigation systems and navigation applications stores information related to a plurality of points, which includes the name of a point (for example, the place name of the point and the name of an object existing at the point) and position information of the point (for example, latitude and longitude, and address), such as POI (Point of Interest) information and information related to a user's home and office. Therefore, a user can search for a route to a destination by inputting the name of the destination into a navigation system or a navigation application (for example, Patent Document 1).

[0003] Japanese Patent Application Laid-Open No. 2002-116039

[0004] However, newly popular spots may not be registered in map data. In order to search for a route to such an unregistered spot, it is necessary to find the spot's position information (for example, the address) through Internet search or the like, which is time-consuming. In addition, spots that were once popular may no longer attract public attention, and objects that were once popular may cease to exist.

[0005] An example of the problem to be solved by the present invention is to keep map data in a useful state.

[0006] In order to solve the above problem, the invention according to claim 1 provides an information processing apparatus comprising: a topicality confirmation processing unit that confirms the topicality of topicality confirmation points, which are at least some of the points registered in map data, using a natural language processing model; and a map data editing processing unit that edits map data based on the topicality of the topicality confirmation points.

[0007] The invention described in claim 17 is an information processing method performed by a computer, comprising: a topicality confirmation processing step that uses a natural language processing model to confirm the topicality of topicality confirmation points, which are at least some of the points registered in map data; and a map data editing processing step that edits map data based on the topicality of the topicality confirmation points.

[0008] The invention described in claim 18 is an information processing program that causes a computer to execute the information processing method described in claim 17.

[0009] The invention described in claim 19 is a computer-readable storage medium that stores the information processing program described in claim 18.

[0010] This figure shows an information processing device 100 according to one embodiment of the present invention. This figure illustrates an example of map data. This figure shows an example of a control unit 110. This figure shows an example of a processing operation performed in the control unit 110 when information regarding the starting point is acquired by the starting point information acquisition processing unit 1101 and the input name is acquired by the input information acquisition processing unit 1102. This figure shows an example of a processing operation performed in the control unit 110 after the user has arrived at an input point candidate (destination point). This figure illustrates the registration of a point corresponding to the input name in the map data. This figure shows an example of a processing operation performed in step S503 of Figure 5. This figure illustrates an example of map data. This figure illustrates the registration of a different name in the map data. This figure shows an example of a processing operation performed in the control unit 110 after the user has arrived at an input point candidate. This figure illustrates the registration of a point corresponding to the input name in the map data. This figure shows an example of a processing operation performed in the control unit 110. This figure shows another example of the control unit 110. This figure shows an example of a processing operation performed in the control unit 110 shown in Figure 13 when a name is entered. This figure shows another example of the control unit 110.

[0011] An information processing device according to one embodiment of the present invention includes a topicality confirmation processing unit that uses a natural language processing model to confirm the topicality of topicality confirmation points, which are at least some of the points registered in the map data, and a map data editing processing unit that edits the map data based on the topicality of the topicality confirmation points. In this embodiment, the map data is edited based on topicality with respect to at least some of the registered points. Therefore, in this embodiment, it is possible to keep the map data in a useful state.

[0012] The map data editing processing unit may also delete locations with low topicality from the map data. The topicality checking processing unit may evaluate the topicality of the locations at a level of 2 or higher, and delete locations with topicality evaluated to be below a predetermined level from the map data. In this way, even if a location no longer exists where it was once a topic of conversation, it will not be deleted from the map data as long as it still has topicality. This makes it possible to accommodate needs such as pilgrimages to sacred sites.

[0013] The map data editing processing unit determines the display format of the topicality-checked location on the map based on the topicality of the location. The topicality-checking processing unit may also evaluate the topicality of the location at two or more levels, and the map data editing processing unit may determine the display format of the location on the map based on the level. In this way, it becomes possible to display the map according to the topicality.

[0014] The topicality verification processing unit may also verify the topicality of the topicality verification points when updating the map data. Doing so makes it possible to keep the map data in a useful state.

[0015] The topicality verification processing unit may change the weighting of information sources used to verify the topicality of a topicality verification location based on predetermined criteria. The topicality verification processing unit may also change the weighting of information sources based on the type of topicality verification location. By doing so, it becomes possible to appropriately evaluate topicality for each type of topicality verification location.

[0016] The topicality verification processing unit may evaluate the level of topicality of a topicality verification location based on the reliability of the information source. The topicality verification processing unit may calculate an index that quantifies the reliability of the information source and evaluate the level of topicality of a topicality verification location based on the calculated index. In this way, it becomes possible to properly evaluate topicality.

[0017] The aforementioned map data editing processing unit may also delete locations whose topicality has rapidly declined from the map data. In this way, it becomes possible to remove spots that have rapidly lost topicality from the map data.

[0018] The map data editing processing unit may edit the map data based on the topicality of the topicality-checked location and the current season. By doing so, it becomes possible to prevent events and spots that are popular for seasonal events or for a limited time from being deleted from the map data when their topicality declines.

[0019] The map data editing processing unit may also determine that locations with low topicality should not be displayed on the map. The map data editing processing unit may also delete locations from the map data if their topicality does not recover within a predetermined period. This prevents locations that have temporarily become less topical from being deleted from the map data.

[0020] The aforementioned map data editing processing unit may also delete locations from the map data if their topicality has remained low for a predetermined period. This prevents locations that have temporarily lost topicality from being deleted from the map data.

[0021] The map data editing processing unit may also determine the display format of the topicality-checked location on the map based on the quality of topicality of that location. In this way, users can recognize the quality of topicality of a location at a glance.

[0022] An information processing method according to one embodiment of the present invention is an information processing method performed by a computer, comprising: a topicality confirmation processing step that uses a natural language processing model to confirm the topicality of topicality confirmation points, which are at least some of the points registered in map data; and a map data editing processing step that edits the map data based on the topicality of the topicality confirmation points. In this embodiment, the map data is edited based on topicality with respect to at least some of the registered points. Therefore, in this embodiment, it is possible to keep the map data in a useful state.

[0023] An information processing program according to one embodiment of the present invention causes a computer to execute the above-described information processing method. Therefore, in this embodiment, it is possible to keep the map data in a useful state.

[0024] A computer-readable storage medium according to one embodiment of the present invention stores the above-mentioned information processing program. Therefore, in this embodiment, the above-mentioned information processing program can be distributed independently in addition to being incorporated into a device, and version upgrades and the like can be easily performed.

[0025] <Information Processing Device 100> Figure 1 shows an information processing device 100 according to one embodiment of the present invention. The information processing device 100 may consist of one device or may consist of multiple devices. The information processing device 100 may be, for example, a portable terminal device carried by a user (e.g., a smartphone), an in-vehicle device installed in a vehicle in which a user is riding (e.g., a navigation system), or a server device that communicates with the portable terminal device or the in-vehicle device. It may also be realized by a combination of a portable terminal device, an in-vehicle device, and a server device.

[0026] The information processing device 100 includes a control unit 110, an input unit 120, a location information acquisition unit 130, a communication unit 140, a storage unit 150, an output unit 160, an imaging unit 170, and an audio acquisition unit 180. If the information processing device 100 is a server device, it does not need to have the input unit 120, the location information acquisition unit 130, the output unit 160, the imaging unit 170, and the audio acquisition unit 180.

[0027] The control unit 110 is an information processing device (computer) for processing information, and includes, for example, a CPU (Central Processing Unit). The input unit 120 is an input device (for example, a touch panel, keyboard, buttons, camera, microphone) for receiving information from the user. The location information acquisition unit 130 is a device for acquiring the location of the information processing device 100, and has, for example, an antenna that receives radio waves transmitted from satellites constituting a GNSS (Global Navigation Satellite System), including GPS (Global Positioning System), and acquires information regarding the current location of a moving object based on the radio waves received by the antenna. The communication unit 140 is a communication device for sending and receiving information with other devices. The storage unit 150 is a storage device (for example, a memory, hard disk drive, solid state drive) for storing information, and stores map data. The output unit 160 is an output device for outputting information, such as a display device for showing information (e.g., a display), an audio output device for outputting audio related to the information (e.g., a speaker), or a printing device for printing information (e.g., a printer). The shooting unit 170 is a shooting device for shooting video (e.g., a camera), and the captured video is stored in the storage unit 150. The audio acquisition unit 180 is a device for capturing audio (e.g., a microphone), and the captured audio is stored in the storage unit 150.

[0028] The map data stored in the memory unit 150 contains multiple locations, and for each of the registered locations, as shown in Figure 2, the name of the location and its location information (address and latitude / longitude) are registered. The name of the location may be, for example, the place name of the location, the common name of the location, or the name or common name of an object located at the location. In the example shown in Figure 2, for a location named "X1", "X1" is registered as the name and "Y1" is registered as the location information, and for a location named "X2", "X2" is registered as the name and "Y2" is registered as the location information.

[0029] Figure 3 shows an example of the control unit 110. It includes a starting point information acquisition processing unit 1101, an input information acquisition processing unit 1102, a destination point setting processing unit 1103, a route search unit 1104, and a route output processing unit 1105.

[0030] The departure point information acquisition processing unit 1101 acquires information about the departure point (departure point information). If the information processing device 100 is a mobile terminal device or an in-vehicle device, the departure point information acquisition processing unit 1101 may, for example, acquire departure point information input from the user via the input unit 120, or it may acquire the current location acquired by the location information acquisition unit 130 as departure point information (location information of the departure point). Also, if the information processing device 100 is a server device, the departure point information acquisition processing unit 1101 may, for example, use communication via the communication unit 140 to acquire departure point information input from the user to the mobile terminal device or in-vehicle device, or it may use communication via the communication unit 140 to acquire the current location of the mobile terminal device or in-vehicle device as departure point information.

[0031] The input information acquisition processing unit 1102 acquires the input name, which is a name entered by the user (for example, the official name or common name of a place the user wants to go, or the official name or common name of something the user wants to see). If the information processing device 100 is a mobile terminal device or an in-vehicle device, the input information acquisition processing unit 1102 acquires the name entered by the user via the input unit 120, for example. If the information processing device 100 is a server device, the input information acquisition processing unit 1102 acquires the name entered by the user via the mobile terminal device or in-vehicle device using communication via the communication unit 140, for example.

[0032] The destination point setting processing unit 1103 sets the destination point based on the input name obtained by the input information acquisition processing unit 1102. If the input name obtained by the input information acquisition processing unit 1102 is included in the names of locations registered in the map data, the destination point setting processing unit 1103 obtains the location corresponding to the input name from the map data and sets the obtained location as the destination point. In other words, the destination point setting processing unit 1103 uses the location information (latitude, longitude, and address) of the location corresponding to the input name obtained from the map data as the location information (latitude, longitude, and address) of the destination point. If the input name obtained by the input information acquisition processing unit 1102 is not included in the names of locations registered in the map data, the destination point setting processing unit 1103 uses a first natural language processing model (for example, a large-scale language model (LLM)) to obtain candidate input locations that are candidates for locations corresponding to the input name, and sets the candidate input location as the destination point. In other words, the destination point setting processing unit 1103 takes the location information (latitude, longitude, and address) of the input location candidate as the location information (latitude, longitude, and address) of the destination point. The destination point setting processing unit 1103 may include a first natural language processing model internally, or it may use an external natural language processing model as the first natural language processing model by using communication via the communication unit 140.

[0033] The location corresponding to the input name is, for example, the location the user wants to go to if the input name is the official name or common name of the place the user wants to go, or, for example, the location where the object corresponding to the input name exists if the input name is the official name or common name of the object the user wants to see.

[0034] The route search unit 1104 performs a route search from the starting point acquired by the starting point information acquisition processing unit 1101 to the destination point set by the destination point setting processing unit 1103.

[0035] The route output processing unit 1105 outputs the route found by the route search unit 1104. If the information processing device 100 is a mobile terminal device or an in-vehicle device, the route output processing unit 1105 outputs the route, for example, by the output unit 160. If the information processing device 100 is a server device, the route output processing unit 1105 outputs the route, for example, by using communication by the communication unit 140, to the output device of the mobile terminal device or in-vehicle device (for example, a display or speaker).

[0036] Therefore, in this embodiment, even if the name entered by the user is not registered in the map data, it is possible to output a route to the location corresponding to the name entered by the user and guide the user.

[0037] Figure 4 shows an example of the processing operations performed in the control unit 110 when the starting point information acquisition processing unit 1101 acquires information about the starting point and the input name is acquired by the input information acquisition processing unit 1102. If the input name is included in the names of locations registered in the map data (step S401, YES), the destination point setting processing unit 1103 acquires the location corresponding to the input name from the map data and sets the acquired location as the destination point (step S402). If the input name is not included in the names of locations registered in the map data (step S401, NO), the destination point setting processing unit 1103 uses a first natural language processing model to acquire candidate input locations that are candidates for locations corresponding to the input name, and sets the candidate input location as the destination point (step S403). The route search unit 1104 performs a route search from the starting point to the destination point set by the destination point setting processing unit 1103 (step S404). The route output processing unit 1105 outputs the route searched by the route search unit 1104 (step S405).

[0038] In this embodiment, if the input name is not included in the names of locations registered in the map data, the destination location setting processing unit 1103 will acquire an input location candidate (a location acquired as a candidate for a location corresponding to the input name using the first natural language processing model), and the user will be guided to this input location candidate. A location different from the location corresponding to the input name may be acquired as an input location candidate. For this reason, it is best to confirm that the input location candidate is the location corresponding to the input name before registering the input location candidate in the map data as the location corresponding to the input name.

[0039] Therefore, in this embodiment, the control unit 110 further includes a user information acquisition processing unit 1106, a user information confirmation processing unit 1107, and a registration processing unit 1108, as shown in Figure 3.

[0040] A user who is guided to a location corresponding to the input name, that is, a user who is guided to the desired location, may acquire information about that location, such as by taking a photograph. Therefore, in this embodiment, the user acquisition information processing unit 1106 acquires user acquisition information, which is information acquired by the user, after the time the user arrives at the input location candidate (destination point) acquired by the destination location setting processing unit 1103. In other words, the user acquisition information processing unit 1106 acquires user acquisition information, which is information acquired by a user who has visited the input location candidate. If the information processing device 100 is a mobile terminal device or an in-vehicle device, the user acquisition information processing unit 1106 may, for example, acquire the current location acquired by the location information acquisition unit 130 as the user's current location, and detect when the user has arrived at the input location candidate based on the acquired current location. Furthermore, if the information processing device 100 is a server device, the starting point information acquisition processing device 1101 may, for example, use communication by the communication unit 140 to acquire the current location of a mobile terminal device or in-vehicle device as the user's current location, and then detect when the user has arrived at a candidate input location based on the acquired current location.

[0041] User-acquired information includes, for example, images taken by the user between the time the user arrives at a candidate input location and the time elapsed for a first period of time. If the user-acquired information includes images taken by the user, the user-acquired information acquisition processing unit 1106 acquires images taken by the imaging unit 170 and stored in the storage unit 150 as images taken by the user if the information processing device 100 is a mobile terminal device or an in-vehicle device, or if the information processing device 100 is a server device, it acquires images taken by the user using communication via the communication unit 140 and stored in the mobile terminal device or in-vehicle device. The first period of time is set as appropriate.

[0042] User-acquired information includes, for example, images uploaded by the user to the cloud between the time the user arrives at a candidate input location and the time elapsed for a second period of time. Images uploaded by the user to the cloud include images posted by the user to so-called SNS (Social Networking Service). If user-acquired information includes images uploaded by the user, the user-acquired information acquisition processing unit 1106 acquires them from the cloud as images uploaded by the user using communication via the communication unit 140. The second period of time is set as appropriate.

[0043] User-acquired information includes, for example, audio recorded by the user between the time the user arrives at a candidate input location and a third time interval has elapsed. If user-acquired information includes audio recorded by the user, the user-acquired information acquisition processing unit 1106 acquires the audio recorded by the audio acquisition unit 180 and stored in the storage unit 150 as audio recorded by the user if the information processing device 100 is a mobile terminal device or an in-vehicle device, or if the information processing device 100 is a server device, it acquires the audio recorded by the mobile terminal device or in-vehicle device and stored using communication by the communication unit 140 as audio recorded by the user. The third time interval is set as appropriate.

[0044] The user acquired information confirmation processing unit 1107 confirms whether information relating to an input name is included in the user acquired information. At this time, for example, the user acquired information confirmation processing unit 1107 uses a second natural language processing model (for example, a large language model (LLM)) to confirm whether information relating to an input name is included in the user acquired information. The second natural language processing model may be the same as the first natural language processing model, or may be different therefrom. The user acquired information confirmation processing unit 1107 may internally include the second natural language processing model, or may use an external natural language processing model as the second natural language processing model via communication performed by the communication unit 140.

[0045] When the user acquired information includes an image captured by the user or an image uploaded by the user, the information relating to the input name includes an image relating to the input name. For example, if the input name is an official name or common name of a point the user wants to go to, the image relating to the input name includes an image captured at the point corresponding to the input name or an image capturing the point corresponding to the input name. If the input name is an official name or common name of an object the user wants to see, the image relating to the input name includes an image capturing the object corresponding to the input name. When the user acquired information includes voice collected by the user, the information relating to the input name includes voice generated at the point corresponding to the input name.

[0046] The user-acquired information confirmation processing unit 1107, when checking whether the user-acquired information includes information about the input name, may use a second natural language processing model to acquire a location related to the user-acquired information, check whether the location related to the user-acquired information and the location corresponding to the input name (candidate input location) indicate the same location, and if the location related to the user-acquired information and the location corresponding to the input name indicate the same location, it may determine that the user-acquired information includes the information about the input name. In this case, for example, if the distance between the location related to the user-acquired information and the location corresponding to the input name is less than or equal to a first distance, the user-acquired information confirmation processing unit 1107 determines that the location related to the user-acquired information and the location corresponding to the input name indicate the same location. The first distance is set as appropriate. If the user-acquired information is an image taken by the user or an image uploaded by the user, the location related to the user-acquired information is, for example, the location where the image was taken, a location shown in the image, or a location where an object shown in the image exists. If the user-acquired information is audio recorded by the user, the location related to the user-acquired information is the location where the audio was generated.

[0047] The presence of information about the input name in the user-acquired information indicates that the user has acquired information about the input name at the location corresponding to the input name, that is, that the user has arrived at the location corresponding to the input name. Therefore, in this embodiment, if the user-acquired information includes information about the input name, the registration processing unit 1108 registers the candidate input location in the map data as the location corresponding to the input name. Specifically, the registration processing unit 1108 registers the input name in the map data as the name of the location corresponding to the input name, and registers the location information of the candidate input location in the map data as the location information of the location corresponding to the input name.

[0048] As described above, in the present embodiment, after confirming that the input point candidate is a point corresponding to the input name, the input point candidate is registered in the map data as the point corresponding to the input name. Therefore, in the present embodiment, no different position information will be registered as the position information of the input name, and from the next time onward, it is possible to accurately guide to the point corresponding to the input name only by inputting the input name. As a result, in the present embodiment, it is possible to keep the map data in a useful state.

[0049] At this time, when registering the input point candidate as the point corresponding to the input name, the registration processing unit 1108 may register the input name and the position information of the point corresponding to the input name in the map data as POI information. By doing this, in the case of a navigation system or a navigation application that displays POI information on a map during route guidance, it becomes possible to display the point corresponding to the registered input name on the map.

[0050] Figure 5 is a diagram showing an example of the processing operation executed by the control unit 110 after the user arrives at the input point candidate (destination point). A user acquired information acquisition processing unit 1106 acquires user acquired information, which is information acquired by the user, after the time when the user arrives at the destination point set by a destination point setting processing unit 1103 (step S501). A user acquired information confirmation processing unit 1107 uses a second natural language processing model to confirm whether information related to the input name is included in the user acquired information (step S502). If information related to the input name is included in the user acquired information (step S502, YES), a registration processing unit 1108 registers the input point candidate as a point corresponding to the input name in the map data (step S503), and ends the processing. If information related to the input name is not included in the user acquired information (step S502, NO), the processing is ended.

[0051] In the processing operation shown in Figure 4, if the name of the object the user wants to see, "XN", is entered as the input name, in step S401, it is checked whether the input name "XN" is registered in the map data. If the input name "XN" is not stored in the map data, in step S403, the location information "YN", which is a candidate input location where "XN" exists, is obtained using the first natural language processing model, and the location information "YN", which is a candidate input location, is set as the location information of the destination point. Then, in the processing operation shown in Figure 5, for example, if an image of "XN" is included in the images taken by the user after arriving at the input location candidate, in step S503, the input location candidate is registered in the map data as the location where "XN" exists, as shown in Figure 6. In other words, as shown in Figure 6, "XN" is registered as the name of the location where "XN" exists, and the location information "YN", which is the location information of the input location candidate, is registered as the location information of the location where "XN" exists.

[0052] <Registration of Alternate Names> Even if a location corresponding to an input name is registered in the map data, a name different from the input name may be registered for that location. Therefore, if the user-acquired information includes information about the input name and the same location as the input location candidate is included in the map data, the registration processing unit 1118 may register the input name as an alternate name for the same location as the input location candidate. In this case, the registration processing unit 1118 should determine that any location within a second distance from the input location candidate is the same location as the input location candidate. The second distance can be set as appropriate and may be the same as the first distance or may be different from the first distance.

[0053] Figure 7 shows an example of the processing operation performed in step S503 of Figure 5. In other words, the processing operation shown in Figure 7 is performed in step S502 of Figure 5 if the user-acquired information includes information about the input name. The registration processing unit 1108 checks whether the same location as the input location candidate is included in the map data (step S701). If the same location as the input location candidate is not included in the map data (step S701, NO), the registration processing unit 1108 registers the input location candidate in the map data as the location corresponding to the input name (step S702). If the same location as the input location candidate is included in the map data (step S701, YES), the input name is registered in the map data as an alternative name for the same location as the input location candidate (step S703).

[0054] The Meganebashi Bridge in Nagasaki City is known as Japan's first stone arch bridge. Furthermore, heart-shaped stones called "heart stones" are embedded in the bridge's embankment, and the bridge is also known as the location of these "heart stones." In the processing operation shown in Figure 4, if the user inputs "heart stones" as the name of the object they want to see, and "heart stones" are not registered in the map data, then in step S403, the location information of the Meganebashi Bridge (32.7472222 degrees North latitude, 129.8800323 degrees East longitude) may be acquired as a candidate location (input location candidate) for where the "heart stones" are located. As a result, the user may be guided to the Meganebashi Bridge. Then, in the processing operations shown in Figures 5 and 7, after arriving at "Meganebashi," the user takes a photo of "Heartstone," and furthermore, regarding the location where "Meganebashi" is located, as shown in Figure 8, if "Meganebashi" is registered as the name in the map data and the location information is registered as "32.7472222 degrees North latitude, 129.8800323 degrees East longitude," then in step S703, as shown in Figure 9, "Heartstone" is registered as an alternative name for "Meganebashi." In other words, as shown in Figure 9, in addition to "Meganebashi," "Heartstone" is registered as the name of the location where the location information is "32.7472222 degrees North latitude, 129.8800323 degrees East longitude."

[0055] <Registered Location Candidate Storage Processing Unit 1109> If a location corresponding to an input name is not registered in the map data, the system may guide the user to the location corresponding to the input name multiple times before registering the location corresponding to the input name in the map data. Therefore, the control unit 110 may further include a registered location candidate storage processing unit 1109.

[0056] If the user-acquired information includes information about the input name and the input location candidate is not already stored in the storage unit 150 as a registration location candidate, the registration processing unit 1109 stores the input location candidate in the storage unit 150 as a registration location candidate. Then, if the user-acquired information includes information about the input name and the input location candidate is already stored in the storage unit 150 as a registration location candidate, the registration processing unit 1108 registers the input location candidate as a location corresponding to the input name in the map data.

[0057] In other words, if the user-acquired information includes information about an input name, and the location corresponding to that input name (candidate input location) is the first location that has been guided, the registered location candidate storage processing unit 1109 stores the input name in the storage unit 150 as the name of the registered location candidate and stores the location of the input location candidate in the storage unit 150. If the location corresponding to that input name (candidate input location) is not the first location that has been guided, the registration processing unit 1108 registers the input name in the map data as the name of the location corresponding to the input name, and registers the location information of the input location candidate in the map data as location information of the location corresponding to the input name.

[0058] The registration processing unit 1108 and the registration location candidate storage processing unit 1109 determine, for example, that if a registration location candidate located within a third distance from the input location candidate is stored in the storage unit 150, then the input location candidate is stored in the storage unit 150 as the registration location candidate. The third distance is set as appropriate and may be the same as the first distance or the second distance, or it may be different from the first distance or the second distance.

[0059] In this case, the names of registered location candidates that are within a third distance from the input location candidate may differ from the input name. Therefore, if the user-acquired information includes information about the input name and the input location candidate is stored in the storage unit 150 as a registered location candidate, the registration processing unit 1108 registers the input name and / or the name for the registered location candidate as the name of the location corresponding to the input name in the map data, and registers the location information of the input location candidate and / or the location information of the registered location candidate as the location information of the location corresponding to the input name in the map data.

[0060] Figure 10 shows an example of processing operations performed in the control unit 110 after the user arrives at a candidate input location. The user acquisition information processing unit 1106 acquires user acquisition information, which is information acquired by the user, after the time the user arrives at the destination location set by the destination location setting processing unit 1103 (step S1001). The user acquisition information confirmation processing unit 1107 uses a second natural language processing model to check whether or not the user acquisition information contains information about the input name (step S1002). If the user acquisition information does not contain information about the input name (step S1002, NO), the process is terminated.

[0061] If the user-acquired information includes information about the input name (step S1002, YES), the registered location candidate storage processing unit 1109 checks whether the input location candidate is stored in the storage unit 150 as a registered location candidate (step S1003). If the input location candidate is not stored in the storage unit 150 as a registered location candidate (step S1003, NO), the registered location candidate storage processing unit 1109 stores the input location candidate in the storage unit 150 as a registered location candidate (step S1004) and terminates the process. If the input location candidate is stored in the storage unit 150 as a registered location candidate (step S1003, YES), the registration processing unit 1108 registers the input location candidate in the map data as a location corresponding to the input name and a location corresponding to the name for the registered location candidate (step S1005) and terminates the process.

[0062] If the user enters "Meganebashi" as the input name for the destination, and "Meganebashi" is not registered in the map data, then in step S403 of the processing operation shown in Figure 4, the location information of "Meganebashi" (32.7472222 degrees North latitude, 129.8800323 degrees East longitude) is acquired as the location information of a candidate location (input location candidate) where "Meganebashi" exists, and as a result, the user is guided to "Meganebashi". Then, in the example processing operation shown in Figure 10, for example, if the user has taken a photograph of "Meganebashi" or a photograph taken from "Meganebashi" among the photographs taken by the user at the input location candidate (the location where "Meganebashi" exists), then in step S1004, "Meganebashi" is stored in the storage unit 150 as the name of the registered location candidate, and "32.7472222 degrees North latitude, 129.8800323 degrees East longitude" is stored as the location information of the registered location candidate. Subsequently, if the user enters "Heartstone" as the input name for the item they want to see, and "Heartstone" is not registered in the map data, then in step S403 of the processing operation shown in Figure 4, the location information of "Meganebashi" (32.7472222 degrees North latitude, 129.8800323 degrees East longitude) will be obtained as the location information of a candidate location (input location candidate) where "Heartstone" exists, and as a result, the user may be guided to "Meganebashi". Then, in the example processing operation shown in Figure 10, for example, if the photos taken by the user after arriving at a candidate input location (the location where "Meganebashi" exists) include an image of "Heartstone", then in step S1005, since the location information of "Meganebashi" (32.7472222 degrees North latitude, 129.8800323 degrees East longitude) is already registered, for example, "Heartstone" and "Meganebashi" are registered in the map data as the names of the locations corresponding to "Heartstone", and the location information of "Meganebashi" (32.7472222 degrees North latitude, 129.8800323 degrees East longitude) is registered in the map data as the location information of the location corresponding to "Heartstone".

[0063] <Deletion of Registered Locations> Spots that are no longer talked about, or spots where the things that were talked about no longer exist, may remain in the map data. Therefore, the control unit 110 of the information processing device 100 according to this embodiment further includes a topicality confirmation processing unit 1110 and a map data editing processing unit 1111.

[0064] The topicality verification processing unit 1110 uses a third natural language processing model (e.g., a large-scale language model (LLM)) to verify the topicality of at least some of the locations registered in the map data (topicality verification locations). The topicality verification locations include, for example, locations registered in the map data by the registration processing unit 1108. The topicality verification processing unit 1110 evaluates the topicality of the topicality verification locations at a level of 2 or higher.

[0065] The map data editing processing unit 1111 edits the map data based on the topicality of the topicality-confirmed locations as confirmed by the topicality-confirming processing unit 1110. If the topicality of the topicality-confirmed locations is evaluated at a level of 2 or higher, the map data editing processing unit 1111 edits the map data based on the topicality level of the topicality-confirmed locations as evaluated by the topicality-confirming processing unit 1110.

[0066] Thus, in this embodiment, map data is edited based on topicality for at least some registered locations. Therefore, in this embodiment, it is possible to keep the map data in a useful state.

[0067] At this time, the map data editing processing unit 1111 deletes from the map data, for example, locations that have been confirmed to be popular by the popularity confirmation processing unit 1110 but have low popularity. For example, if the popularity confirmation processing unit 1110 evaluates the popularity of a location to be at a level of 2 or higher, the map data editing processing unit 1111 deletes from the map data any locations that have been evaluated to be below a predetermined level. In this way, even if a location no longer exists where something that was once popular is no longer popular, it will not be deleted from the map data as long as it still has some popularity. This makes it possible to accommodate needs such as pilgrimages to sacred sites.

[0068] Furthermore, the map data editing processing unit 1111 may determine the display format of the topicality-confirmed location on the map based on the topicality of the topicality-confirmed location confirmed by the topicality-confirming processing unit 1110. When the topicality-confirming processing unit 1110 evaluates the topicality of the topicality-confirmed location at a level of 2 or more, for example, the display format of the topicality-confirmed location on the map is determined based on the topicality level of the topicality-confirmed location evaluated by the topicality-confirming processing unit 1110. In this case, for example, the lower the evaluated topicality level, the fainter and less conspicuous the display on the map may be made, or the lower the evaluated topicality level, the longer the press time required to select it on the map may be made, making it more difficult to select. In this way, it becomes possible to display the map according to the topicality.

[0069] The topicality confirmation processing unit 1110 may, when confirming the topicality of a topicality confirmation point, confirm the topicality for a period from the current time to a point four time prior. The fourth time period may be one week, one month, three months, six months, one year, three years, five years, or the cycle in which the topicality confirmation processing unit 1110 performs processing, and may be set as appropriate.

[0070] Figure 12 shows an example of a processing operation performed by the control unit 110. The processing operation shown in Figure 10 is performed, for example, when map data is updated. The processing operation shown in Figure 12 may be performed at predetermined intervals.

[0071] The topicality verification processing unit 1110 uses a third natural language processing model to verify the topicality of at least some of the locations registered in the map data (topicality verification locations) (step S1201). The map data editing processing unit 1111 edits the map data based on the topicality of the topicality verification locations verified by the topicality verification processing unit 1110 (S1202).

[0072] The topicality verification processing unit 1110 may change the weighting of information sources used to verify the topicality of a topicality verification location based on predetermined criteria. In this case, the topicality verification processing unit 1110 may change the weighting of information sources based on the type of topicality verification location, for example. By doing so, it becomes possible to appropriately evaluate topicality for each type of topicality verification location.

[0073] The types of locations used for checking topicality should include, for example, restaurants, tourist spots, incidents, disasters, festivals, sporting events, and exhibitions. For example, if the type of location is a restaurant, then information from restaurant review sites should be given more weight. If the type of location is a tourist spot, then information from tourist spot review sites should be given more weight, especially regarding the number of social media posts. If there are multiple sites that publish reviews of the location in question, the weighting of each site should be adjusted accordingly.

[0074] The topicality verification processing unit 1110 may evaluate the level of topicality of a topicality verification location based on the reliability of the information source. In this case, the topicality verification processing unit 1110 may, for example, calculate an index that quantifies the reliability of the information source (reliability index) and evaluate the level of topicality of the topicality verification location based on the calculated reliability index. By doing so, it becomes possible to properly evaluate topicality.

[0075] In this case, the topicality verification processing unit 1110 may set the reliability index value so that the reliability index value is high for official websites such as government agency sites, official tourism sites, major news media sites, corporate press releases, and general news sites. The topicality verification processing unit 1110 may also set the reliability index value so that the reliability index value is high for verified SNS accounts. Furthermore, the topicality verification processing unit 1110 may set the reliability index value for personal blogs and SNS posts according to the characteristics of the poster (e.g., number of followers, expertise). Furthermore, the topicality verification processing unit 1110 may set the reliability index value for an information source based on the user behavior of the information source (number of views and shares). For example, the topicality verification processing unit 1110 may set the reliability index value so that the reliability index of an information source with a large number of views or shares is high.

[0076] The map data editing processing unit 1111 may also delete from the map data any locations whose topicality has rapidly decreased among the topics whose topicality has been confirmed by the topicality confirmation processing unit 1110. For example, if the topicality confirmation processing unit 1110 evaluates the topicality of a topicality confirmation location at a level of 3 or higher, the map data editing processing unit 1111 may delete from the map data any topics whose level has fallen below a predetermined level. In this way, it becomes possible to delete spots that have rapidly lost topicality from the map data.

[0077] In the case of seasonal events or spots that become popular for a limited time, their popularity may temporarily decline depending on the season. Therefore, the map data editing processing unit 1111 may edit the map data based on the popularity of the location being checked and the current season. For example, the map data editing processing unit 1111 may edit the map data so that seasonal events are displayed only during the season in which the event is held. Similarly, the map data editing processing unit 1111 may edit the map data so that spots that become popular for a limited time are displayed only during the period in which the spot becomes popular. By doing so, it becomes possible to prevent seasonal events or spots from being deleted from the map data when their popularity declines.

[0078] The map data editing processing unit 1111 may decide that locations with low topicality should not be displayed on the map. Furthermore, if the topicality of a location that has been decided not to be displayed on the map does not recover within a predetermined period, the map data editing processing unit 1111 may delete that location from the map data. In this way, it is possible to prevent locations that have temporarily become low in topicality from being deleted from the map data. In this case, if the topicality checking processing unit 1110 evaluates the topicality of a location at a level of 2 or higher, for example, the map data editing processing unit 1111 may hide locations that have been evaluated as being below a predetermined level from the map data, and then, if the topicality of the hidden location rises above a predetermined level within a predetermined period, it may determine that it has recovered and display the location on the map. If the period during which the topicality of the hidden location is below a predetermined level exceeds a predetermined period, it may determine that it has not recovered and delete the location from the map data. Furthermore, if the map data editing processing unit 1111 receives an instruction from the user to display a location that has been determined not to be displayed on the map, it may also make that location visible on the map.

[0079] The map data editing processing unit 1111 may also delete locations from the map data if their topicality has remained low for a predetermined period. This prevents locations that have temporarily lost topicality from being deleted from the map data.

[0080] The map data editing processing unit 1111 may maintain a history of locations deleted from the map data, and if it receives an instruction from the user to register such locations in the map data, it may register those locations in the map data. In this way, locations that were accidentally deleted or locations that become topics of discussion again will be registered in the map data.

[0081] The map data editing processing unit 1111 may determine the display format of the topicality-checked locations on the map based on the quality of topicality of the locations. For example, the map data editing processing unit 1111 may perform sentiment analysis on the content of the information source (reviews or SNS posts) and use the results of the sentiment analysis as the quality of topicality of the topicality-checked locations. For example, the map data editing processing unit 1111 may change the display format of the topicality-checked locations on the map based on whether the result of the sentiment analysis is positive or negative. For example, if the result of the sentiment analysis is positive, the map data editing processing unit 1111 may highlight the location. The map data editing processing unit 1111 may also display an icon indicating positive sentiment next to locations where the result of the sentiment analysis is positive, and an icon indicating negative sentiment next to locations where the result of the sentiment analysis is negative. In this way, users can recognize the quality of topicality of a location at a glance.

[0082] The map data editing processing unit 1111 may detect trends in the topicality of the topicality confirmation locations and determine the display format of the topicality confirmation locations on the map based on the results of the trend detection. For example, the map data editing processing unit 1111 may detect whether the topicality of the topicality confirmation locations is rapidly increasing or rapidly decreasing, and change the display format of the topicality confirmation locations on the map based on whether it is rapidly increasing or rapidly decreasing.

[0083] The map data editing processing unit 1111 may determine the display format of the topicality confirmation points on the map based on the future topicality of the topicality confirmation points. For example, the map data editing processing unit 1111 may display an icon indicating that an event will be held next to a location where an event will be held in the future. In this case, the map data editing processing unit 1111 may use a fourth natural language processing model (for example, a large-scale language model (LLM)) to detect the future topicality of the topicality confirmation points. In this case, for example, the map data editing processing unit 1111 may detect information about events that will be held in the future at the topicality confirmation points.

[0084] The map data editing processing unit 1111 may determine the display format of the topical locations on the map based on user attributes (e.g., age) and / or user preferences. For example, if the user is a young person, the map data editing processing unit 1111 may highlight popular spots among young people (e.g., general stores), and if the user is a family, it may highlight popular spots among families (e.g., parks).

[0085] <Another Example of the Control Unit 110> Figure 13 shows an alternative example of the control unit 110. The control unit 110 shown in Figure 13 has an input location candidate acquisition processing unit 1112 instead of a departure location information acquisition processing unit 1101, an input information acquisition processing unit 1102, a destination location setting processing unit 1103, a route search unit 1104, and a route output processing unit 1105. If a name is input, the input location candidate acquisition processing unit 1112 uses a first natural language processing model (for example, a large-scale language model (LLM)) to acquire input location candidates that are candidates for locations corresponding to the input name. If the information processing device 100 is a mobile terminal device or an in-vehicle device, the name is input by the input unit 120. If the information processing device 100 is a server device, the name is input to a mobile terminal device or an in-vehicle device that can communicate with the information processing device 100 using communication by the communication unit 140.

[0086] Figure 14 shows an example of the processing operation performed in the control unit 110 shown in Figure 13 when a name is entered. The input location candidate acquisition processing unit 1112 uses a first natural language processing model (for example, a large-scale language model (LLM)) to acquire input location candidates, which are candidates for locations corresponding to the input name (step 1401). The user acquisition information acquisition processing unit 1106 acquires user acquisition information, which is information acquired by a user who visited the input location candidate (step S1402). The user acquisition information confirmation processing unit 1107 checks whether the user acquisition information contains information about the input name (step S1403). If the user acquisition information contains information about the input name (step S1403, YES), the registration processing unit 1108 registers the input location candidate as a location corresponding to the input name in the map data (step S1404), and terminates the process. If the user acquisition information does not contain information about the input name (step S1403, NO), the process terminates.

[0087] In the example of the control unit 110 shown in Figure 13, the user who inputs the input name (first user) and the user who visited the input location candidate (second user) may be the same or different.

[0088] If the first user and the second user are different, it is preferable that the user information acquired by the user information acquisition processing unit 1106 be the information provided by the second user. In this case, for example, the first user (or the user information acquisition processing unit 1106) uploads and presents candidate input locations to the internet (e.g., SNS), and the user information acquisition processing unit 1106 acquires the information provided by the second user in response to this presentation as user information.

[0089] In this case, the user-acquired information confirmation processing unit 1107 may, for example, if the user-acquired information is a photograph, check whether the user-acquired information contains information about the input name based on the location indicated by the geotag of the photograph (the location of the photographing device used to take the photograph when the photograph was taken). For example, the user-acquired information confirmation processing unit 1107 may determine that the user-acquired information contains information about the input name if the location indicated by the geotag is within a predetermined distance from the candidate input location.

[0090] Furthermore, for example, the location history of the terminal used by the second user, the activity history and visit history of the navigation system used by the second user may be acquired from the terminal used by the second user, and the user information acquisition processing unit 1107 may check whether the user information includes information about the input name based on this location history, activity history, and visit history. In this case, the user information acquisition processing unit 1107 may determine that the user information includes information about the input name if the time spent within a predetermined distance from the candidate input location is longer than a predetermined time.

[0091] For example, in a case where "Purin Lab," which was featured on a TV program, is not registered in the map application, the processing operation shown in Figure 14 involves the user entering "Purin Lab" as the name of the place they want to go. In step S1401, information about candidate locations where "Purin Lab" exists (candidate input location) is obtained ("a sweets shop in Shibuya Ward, Tokyo"). Then, information that seems related to "Purin Lab," such as SNS posts made by people who have visited "Purin Lab" (for example, "Purin Lab, which was featured on TV, has opened in Shibuya, so I came!"), is collected (for example, photos). In step S1402, this collected information is obtained as user-acquired information. In step S1403, for example, it is checked whether there are any photos in the user-acquired information whose geotags indicate a location near the candidate input location. If there are photos in the user-acquired information whose geotags indicate a location near the candidate input location, in step S1404, the candidate input location is registered in the map data as a location corresponding to "Purin Lab."

[0092] Furthermore, after obtaining information provided by the second user, the first user (or the user information verification processing unit 1107) may upload this obtained information to the internet (for example, SNS) and present it, allowing users other than the second user to vote on whether or not the information is related to the input name. The user information verification processing unit 1107 may then check, based on the voting results, whether or not the user information contains information related to the input name. When the user information verification processing unit 1107 makes a decision based on the voting results, it may make the decision based on a majority vote, or it may have a fifth natural language processing model (for example, a large-scale language model (LLM)) make the decision based on the voting results.

[0093] If the first user and the second user are the same, the control unit 110 may further include, as shown in Figure 15, a starting point information acquisition processing unit 1101, an input information acquisition processing unit 1102, a destination point setting processing unit 1103, a route search unit 1104, and a route output processing unit 1105, similar to the control unit 110 shown in Figure 3. In this case, if the input name acquired by the input information acquisition processing unit 1102 is not included in the names of locations registered in the map data, the input point candidate acquisition processing unit 1112 uses a first natural language processing model (for example, a large-scale language model (LLM)) to acquire an input point candidate that is a candidate for a location corresponding to the input name. Then, the destination point setting processing unit 1103 sets the input point candidate acquired by the input point candidate acquisition processing unit 1112 as the destination.

[0094] The present invention has been described above with reference to preferred embodiments. Although the present invention has been described with reference to specific examples, various modifications and changes can be made to these examples without departing from the spirit and scope of the invention as described in the claims.

[0095] This application claims priority based on the application "Japanese Patent Application No. 2025-028265" and "PCT / JP2025 / 026968" filed on 25 February 2025, and incorporates all of its disclosures herein.

[0096] 100 Information Processing Unit 110 Control Unit 1101 Departure Point Information Acquisition Processing Unit 1102 Input Information Acquisition Processing Unit 1103 Destination Point Setting Processing Unit 1104 Route Search Unit 1105 Route Output Processing Unit 1106 User Acquisition Information Acquisition Processing Unit 1107 User Acquisition Information Confirmation Processing Unit 1108 Registration Processing Unit 1109 Registered Point Candidate Storage Processing Unit 1110 Topic Relevance Confirmation Processing Unit 1111 Map Data Editing Processing Unit 1112 Input Point Candidate Acquisition Processing Unit 120 Input Unit 130 Location Information Acquisition Unit 140 Communication Unit 150 Storage Unit 160 Output Unit 170 Shooting Unit 180 Audio Acquisition Unit

Claims

1. An information processing device comprising: a topicality verification processing unit that uses a natural language processing model to verify the topicality of topicality verification points, which are at least some of the locations registered in map data; and a map data editing processing unit that edits map data based on the topicality of the topicality verification points.

2. The information processing apparatus according to claim 1, wherein the map data editing processing unit deletes locations that are of low topicality from the map data.

3. The information processing apparatus according to claim 2, wherein the topicality confirmation processing unit evaluates the topicality of the topicality confirmation location at two or more levels, and the map data editing processing unit deletes the topicality confirmation location evaluated to be below a predetermined level from the map data.

4. The information processing apparatus according to claim 1, wherein the map data editing processing unit determines the display form of the topicality confirmation point on the map based on the topicality of the topicality confirmation point.

5. The information processing apparatus according to claim 4, wherein the topicality confirmation processing unit evaluates the topicality of the topicality confirmation location at two or more levels, and the map data editing processing unit determines the display form of the topicality confirmation location on the map based on the levels.

6. The information processing device according to claim 1, wherein the topicality confirmation processing unit confirms the topicality of the topicality confirmation point when the map data is updated.

7. The information processing apparatus according to claim 1, wherein the topicality confirmation processing unit changes the weighting of information sources used to confirm the level of topicality of a topicality confirmation point based on predetermined criteria.

8. The information processing apparatus according to claim 7, wherein the topicality confirmation processing unit changes the weighting of the information source based on the type of topicality confirmation point.

9. The information processing device according to claim 1, wherein the topicality confirmation processing unit evaluates the level of topicality of a topicality confirmation point based on the reliability of the information source.

10. The information processing device according to claim 9, wherein the topicality verification processing unit calculates an index that quantifies the reliability of the information source and evaluates the level of topicality of the topicality verification location based on the calculated index.

11. The information processing apparatus according to claim 1, wherein the map data editing processing unit deletes from the map data locations where the topicality has rapidly decreased.

12. The information processing apparatus according to claim 1, wherein the map data editing processing unit edits the map data based on the topicality of the topicality confirmation location and the current time period.

13. The information processing apparatus according to claim 1, wherein the map data editing processing unit determines that the topicality confirmation locations with low topicality are not displayed on the map.

14. The information processing apparatus according to claim 13, wherein the map data editing processing unit deletes a topicality confirmation point from the map data if the topicality of the topicality confirmation point, which has been determined not to be displayed on the map, does not recover within a predetermined period of time.

15. The information processing apparatus according to claim 1, wherein the map data editing processing unit deletes from the map data locations where the topicality status has remained low for a predetermined period of time.

16. The information processing apparatus according to claim 1, wherein the map data editing processing unit determines the display form of the topicality confirmation point on the map based on the quality of topicality of the topicality confirmation point.

17. An information processing method performed by a computer, comprising: a topicality confirmation processing step that uses a natural language processing model to confirm the topicality of topicality confirmation points, which are at least some of the points registered in map data; and a map data editing processing step that edits map data based on the topicality of the topicality confirmation points.

18. An information processing program that causes a computer to execute the information processing method described in claim 17.

19. A computer-readable storage medium storing the information processing program described in claim 18.

Images