Method of operating an information processing device and a server device

Abstract

To make it possible to obtain higher level information about an area based on GPS coordinates.SOLUTION: An information processing apparatus comprises: a storage unit for storing map image data corresponding to an area and including one or more symbolic representations; and a control unit for embedding spatial coordinates included in the area by machine learning using the map image data and generating a model that estimates first area information corresponding to first spatial coordinates.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present disclosure relates to an information processing apparatus and an operation method of a server apparatus.

Background Art

[0002] A model for embedding GPS (Global Positioning System) coordinates using machine learning is known. For example, Non-Patent Document 1 discloses a technique for training an embedding model using additional semantic data sources such as geotags and geographical / statistical features at positions corresponding to GPS coordinates.

Prior Art Documents

Non-Patent Documents

[0003]

Non-Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] Since conventional GPS coordinate embedding models rely on additional semantic data, they cannot grasp information indicating complex relationships between different locations, which is at a higher level than semantic data.

[0005] The present disclosure relates to an information processing apparatus and the like that enable acquisition of higher-level information of an area including locations based on GPS coordinates.

Means for Solving the Problems

[0006] The information processing device in this disclosure includes a storage unit that stores map image data corresponding to a region and including one or more symbolic representations, and a control unit that uses the map image data to perform machine learning to embed spatial coordinates included in the region and generates a model that estimates first regional information corresponding to a first spatial coordinate.

[0007] The operation method of the server device in this disclosure includes a server device that stores map image data corresponding to a region and including one or more symbolic representations, embedding spatial coordinates included in the region using machine learning with the map image data, and generating a model for estimating first regional information corresponding to a first spatial coordinate. [Effects of the Invention]

[0008] According to the information processing device described in this disclosure, it becomes possible to acquire higher-level information about a region based on GPS coordinates. [Brief explanation of the drawing]

[0009] [Figure 1] This is a diagram showing an example of the configuration of an information processing system. [Figure 2] This is a flowchart illustrating an example of the operation procedure for a server device. [Figure 3] This diagram schematically illustrates machine learning using a server device. [Figure 4] This is a sequence diagram showing an example of the operating procedure of an information processing system. [Modes for carrying out the invention]

[0010] The embodiments will be described below with reference to the drawings.

[0011] Figure 1 shows an example of the configuration of an information provision system in one embodiment. The information provision system 1 has one or more server devices 10 and terminal devices 12, which are connected to each other via a network 11 so as to be able to communicate information with each other.

[0012] The server device 10 is, for example, a server computer belonging to a cloud computing system or other computing system, and functions as a server that implements various functions. The server device 10 may consist of two or more server computers that are connected and operate in cooperation with each other. The server device 10 corresponds to the "information processing device" in this embodiment, which performs information processing for providing various types of information related to a region, for example.

[0013] Terminal device 12 is an information processing device such as a personal computer, smartphone, or tablet terminal, which has communication functions and is configured to perform various information processing tasks. Terminal device 12 is used by the user to obtain various information about a desired region from server device 10.

[0014] Network 11 is, for example, the Internet, but also includes mobile communication networks, ad hoc networks, LANs (Local Area Networks), MANs (Metropolitan Area Networks), or other networks or any combination thereof.

[0015] The server device 10 includes a storage unit 102 that stores map image data corresponding to a region and including one or more symbolic representations, and a control unit 103 that uses the map image data to perform machine learning to embed spatial coordinates included in the region and generates a model for estimating first regional information corresponding to a first spatial coordinate. The map image data is image data in any format and includes schematic representations, i.e., symbolic representations, such as icons and symbols that represent points, facilities, etc., on a map. A region is an area having any shape and area, and can be arbitrarily defined by geographical, economic, or policy factors. Spatial coordinates are coordinates identified by, for example, GPS or other positioning systems, and include, for example, latitude and longitude. Regional information is characteristics of a region and includes various high-level information such as representative land prices, land cover type, population dynamics, and water quality of the region. The server device 10 maps high-level regional information corresponding to each spatial coordinate into a vector space by embedding using the map image data and generates a model (hereinafter referred to as the estimation model) capable of estimating regional information based on the relationships and similarities between spatial coordinates according to the spatial coordinates. Therefore, the server device 10 makes it possible to obtain higher-level information about the region containing an arbitrary spatial coordinate based on that spatial coordinate.

[0016] The configurations of the server device 10 and the terminal device 12 will be described in detail below.

[0017] The server device 10 includes a communication unit 101, a storage unit 102, and a control unit 103. When the server device 10 is composed of two or more server computers, these components are appropriately arranged on two or more computers.

[0018] The communication unit 101 includes one or more communication interfaces. These communication interfaces are, for example, LAN interfaces. The communication unit 101 receives information used in the operation of the server device 10 and transmits information obtained through the operation of the server device 10. The server device 10 is connected to the network 11 via the communication unit 101 and communicates information with the terminal device 12 via the network 11.

[0019] The storage unit 102 includes, for example, one or more semiconductor memories that function as a main memory device, an auxiliary memory device, or a cache memory, one or more magnetic memories, one or more optical memories, or a combination of at least two of these. The semiconductor memory is, for example, a RAM (Random Access Memory) or a ROM (Read Only Memory). The RAM is, for example, a SRAM (Static RAM) or a DRAM (Dynamic RAM). The ROM is, for example, an EEPROM (Electrically Erasable Programmable ROM). The storage unit 102 stores information used for the operation of the server device 10 and information obtained by the operation of the server device 10.

[0020] The control unit 103 includes one or more processors, one or more dedicated circuits, or a combination of these. The processor is, for example, a general-purpose processor such as a CPU (Central Processing Unit), or a dedicated processor such as a GPU (Graphics Processing Unit) specialized for specific processing. The dedicated circuit is, for example, an FPGA (Field-Programmable Gate Array), an ASIC (Application Specific Integrated Circuit), etc. The control unit 103 executes information processing related to the operation of the server device 10 while controlling each part of the server device 10.

[0021] The functions of the server device 10 are realized by a processor included in the control unit 103 executing a control program. The control program is a program for causing a computer to function as the server device 10. Also, some or all of the functions of the server device 10 may be realized by a dedicated circuit included in the control unit 103. Further, the control program may be stored in a non-transitory recording and storage medium readable by the server device 10, and the server device 10 may read it from the medium.

[0022] The memory unit 102 stores map image data corresponding to a region and including one or more symbolic expressions. The map image data is in an arbitrary format such as JPG (Joint Photographic Experts Group) or PNG (Portable Network Graphics). Such map image data is provided from a server that provides map information or is generated based on satellite images or the like. Also, various region information may be stored in the memory unit 102. For example, land price information, vital statistics, land cover type, water quality, etc. are acquired from a server that provides each information and stored.

[0023] The terminal device 12 includes a communication unit 121, a memory unit 122, a control unit 123, a positioning unit 124, an input unit 125, and an output unit 126.

[0024] The communication unit 121 includes a communication module corresponding to a wired or wireless LAN standard, a module corresponding to a mobile communication standard such as LTE (Long Term Evolution), 4G (Fourth Generation), or 5G (Fifth Generation), etc. The terminal device 13 is connected to the network 11 via the communication unit 121 through a nearby router device or a base station of mobile communication, and performs information communication with other devices via the network 11.

[0025] The memory unit 122 includes one or more semiconductor memories, one or more magnetic memories, one or more optical memories, or a combination of at least two of these. The semiconductor memory is, for example, a RAM or a ROM. The RAM is, for example, a SRAM or a DRAM. The ROM is, for example, an EEPROM. The memory unit 122 functions as, for example, a main memory device, an auxiliary memory device, or a cache memory. The memory unit 122 stores information used for the operation of the control unit 123 and information obtained by the operation of the control unit 123.

[0026] The control unit 123 has, for example, one or more general-purpose processors such as a CPU or MPU (Micro Processing Unit), or one or more dedicated processors specialized for a specific process. Alternatively, the control unit 123 may have one or more dedicated circuits such as FPGAs or ASICs. The control unit 123 comprehensively controls the operation of the terminal device 13 by operating according to a control and processing program, or by operating according to an operating procedure implemented as a circuit. The control unit 123 then sends and receives various information with the server device 10, etc., via the communication unit 121 and executes the operations according to this embodiment.

[0027] The positioning unit 124 includes one or more GNSS (Global Navigation Satellite System) receivers. GNSS includes, for example, at least one of GPS, QZSS (Quasi-Zenith Satellite System), BeiDou, GLONASS (Global Navigation Satellite System), and Galileo. The positioning unit 124 acquires location information from the terminal device 12 and sends it to the control unit 123.

[0028] The input unit 125 includes one or more input interfaces. The input interfaces include, for example, physical keys, capacitive keys, pointing devices, touchscreens integrated with a display, cameras for capturing images or image codes, or IC card readers. The input interfaces may also include microphones for receiving voice input. The input unit 125 receives input of information used for the operation of the control unit 123 and sends the input information to the control unit 123.

[0029] The output unit 126 includes one or more output interfaces. The output interfaces are, for example, a display or a speaker. The display is, for example, an LCD or an organic EL display. The output unit 126 outputs information obtained by the operation of the control unit 123.

[0030] The functions of the control unit 123 are realized by the execution of a control program by the processor included in the control unit 123. The control program is a program that causes the processor to function as the control unit 123. In addition, some or all of the functions of the control unit 123 may be realized by dedicated circuits included in the control unit 123.

[0031] Figure 2 is a flowchart illustrating the procedure by which the server device 10 generates an estimation model. The steps in Figure 2 are executed by the control unit 103. The procedure in Figure 2 is executed, for example, in response to instructions from the terminal device 12. For example, when an operator uses the terminal device 12 to send an instruction to the server device 10 to generate an estimation model, the control unit 103 of the server device 10 starts the procedure in Figure 2 in response to that instruction.

[0032] In step S21, the control unit 103 acquires spatial coordinates. The control unit 103 reads a set of spatial coordinates that is pre-stored in the storage unit 102. The set of spatial coordinates is acquired and stored in the storage unit 102, for example, from another server that provides map information, or by extracting it from the provided map information.

[0033] In step S22, the control unit 103 acquires map image data. The control unit 103 reads map image data that has been pre-stored in the storage unit 102. Map image data is acquired and stored in the storage unit 102, for example, from a server that provides map information, or by extracting it from provided map information. One piece of map image data is, for example, an image that shows a map of a region corresponding to an arbitrary shape such as a rectangle and an arbitrary area, and includes a symbolic representation of that region. The control unit 103 acquires any number of map image data.

[0034] In step S23, the control unit 103 filters the map image data. The control unit 103 filters the map image data according to an arbitrary criterion. The arbitrary criterion is, for example, a criterion for the magnitude of the entropy of the map image, and the control unit 103 excludes map image data whose association with the set of spatial coordinates is lower than the criterion.

[0035] In step S24, the control unit 103 extracts spatial coordinates from the set of spatial coordinates for use in machine learning. For example, the control unit 103 extracts spatial coordinates whose coordinate distance from spatial coordinates already included in the map image data is less than or equal to an arbitrary criterion. If an estimation model has not already been generated, step S24 may be omitted.

[0036] In step S25, the control unit 103 performs machine learning using the filtered map image data and embeds spatial coordinates.

[0037] Figure 3 is a schematic diagram illustrating the machine learning process in step S25. The control unit 103 inputs map image data 31 corresponding to the set of spatial coordinates 30 into the teacher network 34 as teacher input data 32, causing the teacher network 34 to generate teacher embedding data 36. The control unit 103 also inputs a subset of the teacher input data 32 into the learning network 35 as learning input data 33, causing the learning network 35 to generate learning embedding data 37. The control unit 103 then causes the learning network 35 to perform self-supervised learning such that the loss 39 between the teacher embedding data 37 and the learning embedding data 27 is minimized. The spatial coordinates 30 used in self-supervised learning are associated with regional information 38, which includes characteristics such as land price, land cover type, population dynamics, and water quality of the corresponding region. Therefore, the learning embedding data 37 includes regional information 300 corresponding to the spatial coordinates 30. In this way, an estimation model is generated that can estimate regional information 300 corresponding to any spatial coordinate 30.

[0038] In a modified version of step S25, the control unit 103 causes the learning network 35 to generate map image data 331 by reconstructing the map image corresponding to the spatial coordinates 30, thereby generating learning-side embedded data 37. Furthermore, the control unit 103 causes the teacher network 34 to cluster the teacher-side embedded data 36 into clusters corresponding to the spatial coordinates 30, and the learning network 35 to cluster the learning-side embedded data 37 into clusters corresponding to the spatial coordinates 30. Then, the control unit 103 causes the learning network 35 to perform self-supervised learning such that the loss 39 between the clusters of the teacher-side embedded data 37 and the clusters of the learning-side embedded data 27 is minimized.

[0039] Figure 4 is a sequence diagram illustrating the coordinated operation of the server device 10 and the terminal device 12 in an embodiment of the information provision system 1. Here, the server device 10 is a server computer having the estimation model shown in Figure 3, or a server computer that can communicate with a server computer having the estimation model and make the estimation model available.

[0040] In step S40, the terminal device 12 sends information specifying spatial coordinates and the type of regional information to the server device 10. The control unit 123 receives input operations from the user specifying desired spatial coordinates and the type of regional information via the input unit 125. For example, the control unit 123 displays a map image acquired from the server device 10, etc., via the output unit 126 and accepts operations from the user specifying a desired location and specifying the desired type of regional information. The type of regional information is one or more of the characteristics of the region corresponding to the specified spatial coordinates, such as land value, land cover type, population dynamics, and water quality. The control unit 123 sends information specifying the spatial coordinates to the server device 10 via the communication unit 121. In the server device 10, the control unit 103 receives information sent from the terminal device 12 via the communication unit 101.

[0041] In step S41, the server device 10 estimates regional information corresponding to the specified spatial coordinates and type of regional information. The control unit 103 uses the estimation model to estimate regional information for the area corresponding to the spatial coordinates.

[0042] In step S42, the server device 10 sends the estimated regional information to the terminal device 12. The control unit 103 sends the estimated regional information to the terminal device 12 via the communication unit 101. In the terminal device 12, the control unit 123 receives the information sent from the server device 10 via the communication unit 121.

[0043] In step S43, the terminal device 12 outputs regional information. The control unit 123 displays or outputs the regional information to the user via the output unit 126. This allows the user to obtain the desired regional information.

[0044] According to the procedure described above, users can obtain higher-level information about the region containing any given spatial coordinates.

[0045] According to the information provision system 1 of this embodiment, it is possible to support decision-making in urban development, construction, and real estate development by estimating regional information such as land prices and population dynamics. Furthermore, by estimating regional information such as land cover type and water quality, it is possible to support decision-making in agricultural monitoring, precision agriculture, and water supply management during natural disasters.

[0046] Furthermore, the server device 10 may collect location information, activity logs, comments, and life logs provided by users using the terminal device 12 via SNS (Social Network Service), etc., and use them as regional information added to map image data to generate an estimation model. According to such an estimation model, for example, it becomes possible to estimate the cost performance of restaurants, etc., in a region that includes spatial coordinates specified by the user, such as the current location or destination. It also becomes possible to estimate pedestrian flow in any spatial coordinate in smart cities, etc. Moreover, this embodiment can also be applied to a service that estimates advertising information suitable for the region that includes the user's current location and travel and tourism information suitable for the region that includes the user's destination, and provides the estimated information to the user.

[0047] In the above embodiment, the processing and control program that defines the operation of the control unit 123 of the terminal device 12 may be stored in the storage unit of the server device 10 or other server device and downloaded to each terminal device 12 via the network 11, or it may be stored in a recording and storage medium readable by each terminal device 12 and read by each terminal device 12 from the medium.

[0048] As described above, embodiments have been explained based on various drawings and examples, but it should be noted that those skilled in the art will find it easy to make various modifications and alterations based on this disclosure. Therefore, it should be noted that these modifications and alterations are within the scope of this disclosure. For example, the functions, etc., included in each means, each step, etc., can be rearranged in a logically consistent manner, and multiple means, steps, etc., can be combined into one or divided. [Explanation of Symbols]

[0049] 1. Information Processing System 10 Server devices 11 Network 12 Terminal devices 101, 121 Communications Department 102, 122 Storage section 103, 123 Control Unit 124 Positioning Unit 125 Input section 126 Output section

Claims

1. A storage unit that stores map image data corresponding to a region and including one or more symbolic representations, A control unit that uses the aforementioned map image data to perform machine learning to embed spatial coordinates included in the region and generates a model for estimating first regional information corresponding to the first spatial coordinates, It has, The control unit, as part of the machine learning process, causes the teacher network to generate teacher-side embedded data using teacher-side input data generated based on the map image data, and causes the learning network to generate learning-side embedded data using a subset of the teacher-side input data as learning-side input data, and then performs self-supervised learning based on the teacher-side embedded data and the loss of the learning-side embedded data to generate a model for estimation. Information processing device.

2. In claim 1, The control unit causes the learning network to generate the learning-side embedded data using the features corresponding to the spatial coordinates as the regional information, and then executes the self-supervised learning. Information processing device.

3. In claim 2, The aforementioned features are one or more of the following: land price, land cover type, population dynamics, water quality indicators, and the spatial coordinates themselves, corresponding to the spatial coordinates. Information processing device.

4. In claim 1, The control unit instructs the learning network to reconstruct map image data corresponding to the spatial coordinates based on the learning-side embedded data, and to perform the self-supervised learning. Information processing device.

5. In claim 1, The control unit causes the teacher network to determine a first cluster corresponding to the map image data based on the teacher-side embedded data, the learning network to determine a second cluster corresponding to the map image data based on the learning-side embedded data, and the self-supervised learning to be performed based on the losses derived from the first and second clusters.

6. An information processing device that can utilize the model described in claim 2, A communication unit that communicates with the terminal device, The terminal device has a control unit that estimates the features of the location specified by the user based on the spatial coordinates of the location specified by the user using the model, and sends information indicating the features to the terminal device. Information processing device.

Images