Position Determination Method, Device, Electronic Device, and Storage Medium

Abstract

The present disclosure provides a location determination method, apparatus, electronic device, and storage medium. The method includes: by obtaining the user address text information associated with multiple login information and the positioning location information of the client logged in with the multiple login information, determining the correspondence between the user address text information and the positioning location information according to the login information, the user address text information, and the positioning location information, and thus determining the location identified by the user address text information according to the correspondence between the user address text information and the positioning location information. Since the user address text information is associated with the login information, and the positioning location information obtained is also the positioning location information of the client logged in with the login information, it is possible to determine the location identified by the user address text information without manual collection, thereby solving the technical problem in the related art that location information needs to be collected manually.

Description

Technical Field

[0001] The present disclosure relates to the technical field of positioning, and in particular, to a method, apparatus, electronic device, and storage medium for determining a location. Background Art

[0002] Geocoding is a coding method based on spatial positioning technology, which is a technology for converting address description text into geographic coordinates. To build a geocoding system, a "geocoding database" needs to be established first. Currently, in related technologies, the "geocoding database" shown below is mostly used to build a geocoding system. As Figure 1 shown, when a user inputs the place name "Building X, XX Community, B District, A City" to query the longitude and latitude of Building X in XX Community, B District, A City. Since this address has been included in the "geocoding database", the geocoding system can successfully complete the matching and return the geographic coordinates of Building X in XX Community, B District, A City to the user. Figure 1 Therefore, to establish a "geocoding database", rich and accurate "place name - longitude and latitude" mapping entries need to be included. In related technologies, the "place name - longitude and latitude" mapping entries are mainly included by manual annotation, that is, hiring and training personnel to record place names and their corresponding longitudes and latitudes offline. In addition, map service providers can also enrich the entries through the "crowdsourcing" method. For example, in a map software, users can voluntarily upload the place names they are familiar with and select the location of the place name in the software to record the longitude and latitude.

[0003] However, the method of hiring manual labor is very labor - and cost - consuming, and the collection period is very long, with a large update delay. In addition, it is inconvenient to enter enclosed plots such as residential communities, and it is difficult to include the location of specific building numbers in the community. The crowdsourcing collection method depends on the active operation of users and requires encouraging users to participate voluntarily.

[0004] The present disclosure aims to at least solve one of the technical problems in the related technologies to some extent. Summary of the Invention

[0005]

[0006] For this purpose, the first object of the present disclosure is to propose a location determination method to realize the correspondence between the user address text information associated with multiple login information and the positioning location information of the client on which the user logs in, so as to determine the location identified by the user address text information according to the correspondence between the user address text information and the positioning location information, and solve the technical problem of collecting location information by manpower in the related technologies.

[0007] The second object of the present disclosure is to propose a location determination apparatus.

[0008] The third object of the present disclosure is to provide an electronic device.

[0009] The fourth object of the present disclosure is to provide a non-transitory computer-readable storage medium.

[0010] The fifth object of the present disclosure is to provide a computer program product.

[0011] It should be noted that in the technical solution of the present disclosure, the acquisition, storage, application, etc. of the user's personal information all comply with the provisions of relevant laws and regulations and do not violate public order and good customs.

[0012] To achieve the above object, an embodiment of the first aspect of the present disclosure provides a method, including:

[0013] Obtaining user address text information associated with a plurality of login information;

[0014] Obtaining positioning location information of clients logged in with the plurality of login information;

[0015] Determining a correspondence relationship between the user address text information and the positioning location information according to the login information, the user address text information, and the positioning location information;

[0016] Determining a location identified by the user address text information according to the correspondence relationship between the user address text information and the positioning location information.

[0017] Optionally, as a first possible implementation manner of the first aspect, the location identified by the user address text information includes the range of an identified area;

[0018] The determining the location identified by the user address text information according to the correspondence relationship between the user address text information and the positioning location information includes:

[0019] Determining, according to the correspondence relationship, positioning location information that matches the name of the identified area from the plurality of positioning location information;

[0020] Determining the range of the identified area according to the positioning location information that matches the name of the identified area.

[0021] Optionally, as a second possible implementation manner of the first aspect, the determining the range of the identified area according to the positioning location information that matches the name of the identified area includes:

[0022] Adjusting the position of a selection box until the number of positioning location information within the selection box is maximized;

[0023] Determine the range of the identification area according to the coverage range of the selection box when the quantity of the positioning position information within the selection box is maximized.

[0024] Optionally, as a third possible implementation manner of the first aspect, the positions indicated by the user address text information further include the positions of multiple objects within the identification area; after determining the range of the identification area according to the positioning position information matching the name of the identification area, it further includes:

[0025] For any one object, determine the position of the corresponding object according to the positioning position information matching the name of the object;

[0026] Generate a position sequence according to the positions of the multiple objects;

[0027] Use a recurrent neural network to correct the positions of the position sequence to obtain an output corrected position sequence;

[0028] Correct the corresponding positions in the position sequence according to the correction data in the corrected position sequence.

[0029] Optionally, as a fourth possible implementation manner of the first aspect, the use of a recurrent neural network to correct the positions of the position sequence to obtain an output corrected position sequence includes:

[0030] Input each position in the position sequence into the recurrent neural network of the encoder for encoding to obtain the hidden state corresponding to each position in the position sequence;

[0031] Input the hidden states corresponding to each position in the position sequence into the recurrent neural network of the decoder for decoding to obtain the corresponding correction data in the corrected position sequence.

[0032] Optionally, as a fifth possible implementation manner of the first aspect, the step of, for any one object, determining the position of the corresponding object according to the positioning position information matching the name of the object includes:

[0033] For any one object, if there are multiple pieces of positioning position information matching the name of the object, determine the position of the object according to the centroid position of the multiple pieces of positioning position information.

[0034] The location determination method proposed in the embodiments of the present disclosure realizes determining the correspondence between the user address text information and the positioning location information according to the login information, the user address text information, and the positioning location information by obtaining the user address text information associated with multiple login information and the positioning location information of the clients logged in by the multiple login information. Thus, according to the correspondence between the user address text information and the positioning location information, the location identified by the user address text information is determined. Since the user address text information is associated with the login information, and the positioning location information is also the positioning location information of the clients logged in by the login information, it is possible to determine the location identified by the user address text information without manual collection, solving the technical problem of needing to collect location information manually in the related art.

[0035] To achieve the above object, an embodiment of the second aspect of the present disclosure proposes a location determination device, including:

[0036] A first acquisition module, configured to acquire the user address text information associated with multiple login information;

[0037] A second acquisition module, configured to acquire the positioning location information of the clients logged in by the multiple login information;

[0038] A determination module, configured to determine the correspondence between the user address text information and the positioning location information according to the login information, the user address text information, and the positioning location information;

[0039] A positioning module, configured to determine the location identified by the user address text information according to the correspondence between the user address text information and the positioning location information.

[0040] Optionally, as the first possible implementation manner of the second aspect, the location identified by the user address text information includes the range of the identified area; the positioning module includes:

[0041] A first determination unit, configured to determine, according to the correspondence, the positioning location information that matches the name of the identified area from the multiple positioning location information;

[0042] A second determination unit, configured to determine the range of the identified area according to the positioning location information that matches the name of the identified area.

[0043] Optionally, as the second possible implementation manner of the second aspect, the second determination unit is configured to:

[0044] Adjust the position of the selection box until the number of positioning location information within the selection box is maximized;

[0045] Determine the range of the identification area according to the coverage range of the selection box when the quantity of the positioning position information within the selection box is maximized.

[0046] Optionally, as a third possible implementation manner of the second aspect, the position identified by the user address text information further includes the positions of multiple objects within the identification area; the apparatus further includes a correction module, including:

[0047] A third determination unit, configured to determine the position of a corresponding object according to the positioning position information matching the name of the object for any one object;

[0048] A generation unit, configured to generate a position sequence according to the positions of the multiple objects;

[0049] A first correction unit, configured to perform position correction on the position sequence by using a recurrent neural network to obtain an output corrected position sequence;

[0050] A second correction unit, configured to correct the corresponding positions in the position sequence according to the correction data in the corrected position sequence.

[0051] Optionally, as a fourth possible implementation manner of the second aspect, the first correction unit is configured to:

[0052] Input each position in the position sequence into the recurrent neural network of the encoder for encoding to obtain the hidden state corresponding to each position in the position sequence;

[0053] Input the hidden state corresponding to each position in the position sequence into the recurrent neural network of the decoder for decoding to obtain the corresponding correction data in the corrected position sequence.

[0054] Optionally, as a fifth possible implementation manner of the second aspect, the third determination unit is configured to:

[0055] For any one object, if there are multiple positioning position information matching the name of the object, determine the position of the object according to the centroid position of the multiple positioning position information.

[0056] The location determination device proposed in the embodiments of the present disclosure obtains the user address text information associated with multiple login information and the positioning location information of the clients logged in with the multiple login information, and realizes determining the correspondence between the user address text information and the positioning location information according to the login information, the user address text information, and the positioning location information, so as to determine the location identified by the user address text information according to the correspondence between the user address text information and the positioning location information. Since the user address text information is associated with the login information, and the positioning location information is also the positioning location information of the clients logged in with the login information, it is possible to determine the location identified by the user address text information without manual collection, solving the technical problem of needing to collect location information manually in the related art.

[0057] To achieve the above object, an embodiment of the third aspect of the present disclosure proposes an electronic device, including:

[0058] At least one processor; and

[0059] A memory communicatively connected to the at least one processor; wherein,

[0060] The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the method described in the first aspect.

[0061] To achieve the above object, an embodiment of the fourth aspect of the present disclosure proposes a non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to cause the computer to execute the method described in the first aspect.

[0062] To achieve the above object, an embodiment of the fifth aspect of the present disclosure proposes a computer program product, including a computer program, and the computer program realizes the method described in the first aspect when executed by a processor.

[0063] Additional aspects and advantages of the present disclosure will be given in part in the following description, become apparent in part from the following description, or be learned through the practice of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] The above and / or additional aspects and advantages of the present disclosure will become apparent and be readily understood from the following description of the embodiments in conjunction with the drawings, where:

[0065] Figure 1 It is a schematic diagram of the principle of determining the geographic coordinates corresponding to the address text based on the geocoding database in the related art;

[0066] Figure 2 It is a schematic flowchart of a location determination method provided by an embodiment of the present disclosure;

[0067] Figure 3 A framework diagram for determining the correspondence between user address text information and positioning location information provided by an embodiment of the present disclosure;

[0068] Figure 4 A flowchart of another location determination method provided by an embodiment of the present disclosure;

[0069] Figure 5 A flowchart of a method for determining the position of a region selection box by brute-force enumeration provided by an embodiment of the present disclosure;

[0070] Figure 6 A flowchart of another location determination method provided by an embodiment of the present disclosure;

[0071] Figure 7 A flowchart of the correction process of a recurrent neural network provided by an embodiment of the present disclosure;

[0072] Figure 8 A flowchart of the rough positioning of building coordinates provided by an embodiment of the present disclosure;

[0073] Figure 9 A flowchart of the fine-tuning of building coordinates provided by an embodiment of the present disclosure;

[0074] Figure 10 A structural diagram of a location determination device provided by an embodiment of the present disclosure;

[0075] Figure 11 A structural diagram of another location determination device provided by an embodiment of the present disclosure; and

[0076] Figure 12 A structural diagram of an electronic device provided by an embodiment of the present disclosure. Detailed implementation manners

[0077] The embodiments of the present disclosure will be described in detail below. The examples of the embodiments are shown in the accompanying drawings, where the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions from beginning to end. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to explain the present disclosure, but should not be construed as limiting the present disclosure.

[0078] It should be noted that in the technical solutions of the present disclosure, the acquisition, storage, and application of the user's personal information involved all comply with the provisions of relevant laws and regulations and do not violate public order and good customs.

[0079] The location determination method, device, electronic device, and storage medium of the embodiments of the present disclosure will be described below with reference to the accompanying drawings.

[0080] Figure 2 The flowchart of a location determination method provided by an embodiment of the present disclosure.

[0081] In the related art, location information is mainly obtained by manual collection. However, the manual method is very labor-intensive and costly, and the collection period is very long, with a large update delay. In addition, it is inconvenient to enter enclosed areas such as residential communities, and it is difficult to include the location of specific building numbers in the community.

[0082] To solve this problem, an embodiment of the present disclosure provides a location determination method to implement the correspondence between the user address text information associated with multiple login information and the positioning location information of the client on which the user logs in, so as to determine the location identified by the user address text information according to the correspondence between the user address text information and the positioning location information, and solve the technical problem of manually collecting location information in the related art.

[0083] As Figure 2 shown, the location determination method includes the following steps:

[0084] Step 201: Obtain the user address text information associated with multiple login information.

[0085] It should be noted that the location determination method provided in this embodiment can be executed by a location determination device. The location determination device can be an electronic device or can be configured in an electronic device to determine the location identified by the user address text information based on the correspondence between the user address text information and the positioning location information.

[0086] Among them, the electronic device can be any stationary or mobile computing device capable of data processing, such as mobile computing devices such as laptop computers and smart phones, or stationary computing devices such as desktop computers, or servers, or other types of computing devices, etc. There is no limitation in this embodiment.

[0087] In this embodiment, the login information can be understood as the information for a user to log in to any client, and there is a one-to-one correspondence with the user, that is, the corresponding user can be determined according to the login information. Optionally, the user address text information associated with the login information can be the user's delivery address text information shared by the user corresponding to the login information. For example, when a user uses an online shopping software to shop online, the user needs to log in to the personal account and fill in the delivery address. In this way, the online shopping software stores a large amount of login information corresponding to the user and the user address text information associated with the login information. Thus, with the consent of the user himself / herself, the user address text information associated with the login information stored in the online shopping software can be obtained through network transmission or physical copy, that is, the user address text information associated with multiple login information can be obtained.

[0088] It should be noted that the location determination device in this embodiment can obtain the user address text information associated with the login information through various public, legal, and compliant methods. For example, after obtaining the authorization of the user, the location determination device can collect in real time the user address text information associated with the login information when the user logs in, or can also obtain the user address text information associated with the login information from other devices after obtaining the authorization of the user corresponding to the login information, or can also obtain the user address text information associated with the login information through other public, legal, and compliant methods. This embodiment does not limit this.

[0089] It can be understood that after obtaining the user address text associated with multiple login information, the obtained user address text information associated with multiple login information can be stored in the user address database.

[0090] Step 202, obtain the positioning location information of the clients logged in by multiple login information.

[0091] Optionally, the positioning location information of the client logged in by the login information can be the user's GPS (Global Positioning System) location information shared by the user corresponding to the login information. For example, when a user uses a mobile terminal such as a mobile phone to log in to the client, the client can send a permission request to obtain the GPS location information of the mobile terminal to the user. Thus, when the user consents to the permission request, the positioning location information of the client logged in by the login information corresponding to the user can be obtained by obtaining the GPS location information of the mobile terminal.

[0092] Similarly, the location determination device in this embodiment can also obtain the positioning location information of the client logged in with the login information through various public, legal, and compliant methods. For example, after obtaining user authorization, the location determination device can collect the positioning location information of the client logged in with the login information in real time when the user logs in, or can also obtain the positioning location information of the client logged in with the login information from other devices after obtaining the authorization of the user corresponding to the login information, or can also obtain the user address text information associated with the login information through other public, legal, and compliant methods. This embodiment does not limit this.

[0093] It should be noted that the "multiple login information" in this step and the "multiple login information" in the previous step should be the same login information, that is, for the same multiple login information, the user address text information associated with the login information and the positioning location information of the client logged in with the login information are obtained respectively.

[0094] Similarly, after obtaining the positioning location information of the clients logged in with multiple login information in this step, the positioning location information of each client logged in with the login information obtained can be stored in the positioning location database.

[0095] Step 203: Determine the correspondence between the user address text information and the positioning location information according to the login information, the user address text information, and the positioning location information.

[0096] Since the user address text information associated with the login information indicates one of the possible permanent residences of the user, and the positioning location information of the client logged in with the login information indicates the GPS location point of the user, there will be a correspondence between the user address text information associated with the same login information and the positioning location information of the client logged in with the login information. Based on this, the location identified by the user address text information can be determined. For example, when the number of users of the user address text information is large enough, if a certain location is covered by the GPS location points of a large number of users, then it is very likely that this location is the location identified by the user address text information. Specifically, for example, if the user address text information associated with multiple login information is all "Building 6, XX Community", then the positioning location information of the clients logged in with multiple login information will also mostly appear near "Building 6, XX Community", so that the location identified by the user address text information associated with multiple login information can be determined.

[0097] In this embodiment, after obtaining the user address text information associated with multiple login information and the positioning location information of the clients logged in by the multiple login information, since there is a corresponding relationship between the user address text information associated with the same login information and the positioning location information of the client logged in by the login information, the corresponding relationship between the user address text information and the positioning location information can be determined based on the login information, the user address text information, and the positioning location information.

[0098] To clearly illustrate how this embodiment determines the corresponding relationship between the user address text information and the positioning location information based on the login information, the user address text information, and the positioning location information, this embodiment provides Figure 3 the framework diagram for determining the corresponding relationship between the user address text information and the positioning location information as shown in Figure 3 As shown, determining the corresponding relationship between the user address text information and the positioning location information can be divided into two stages:

[0099] The first stage is Figure 3 the user association process shown. Given any user address text information, all users who filled in the user address text can be found from the user address database, thus obtaining the corresponding user list. Taking the given user address text information as address A as an example, since the delivery addresses of users 1, 2, and 3 all filled in address A, the corresponding user list obtained is {1, 2, 3}. As a possible implementation, the corresponding user list can be obtained by matching the address text string.

[0100] The second stage is Figure 2 the positioning acquisition process shown. According to the obtained user list, combined with the positioning location database, by extracting the corresponding positioning location information in the positioning location database, the positioning location information corresponding to the user address text information of each user in the user list can be determined. As a possible implementation, the corresponding positioning location information in the positioning location database can be extracted using the user ID (Identity Document, unique identification code) as the keyword. This stage further obtains the positioning location information of multiple users corresponding to each user address text information, completing the process from the user address text information to the positioning location information, that is, A→{1,2,3}→{the positioning location of 1, the positioning location of 2, the positioning location of 3}, thus determining the corresponding relationship between the user address text information and the positioning location information. Taking the user address text information as address A as an example, the final distribution of the user positioning locations corresponding to address A is as shown in Figure 3 the upper right corner.

[0101] In summary, according to the user address text information, the user lists corresponding to each place name can be extracted, with one place name corresponding to one user list. Then, based on the location information of each user in the user list, the user location information corresponding to each user address text information can be determined. Among them, the user address text information can be a POI (Point Of Interest, which generally refers to a series of geographical entities containing multiple buildings such as communities and office areas in this disclosure) place name, such as "xx Community", "xx Office Park", etc. Alternatively, optionally, the user address text information can be a place name containing a POI name and a building number, such as "Building x in xx Community", etc. This embodiment does not limit this.

[0102] It should be noted that since location information is extremely sensitive personal information of users, in order to protect the personal information of users and avoid leakage, the user IDs obtained in this step need to be desensitized and encrypted. As a possible implementation method, encryption can be achieved through, for example, the MD5 (Message-Digest Algorithm 5) algorithm. In this way, it can be ensured that the user information accessed in this embodiment is all anonymous users, and it is difficult to crack the detailed location information of specific users.

[0103] Step 204: Determine the location identified by the user address text information according to the correspondence between the user address text information and the location information.

[0104] In this embodiment, based on the correspondence between the user address text information and the location information determined in the previous step, the location identified by the user address text information can be determined. As a possible implementation method, according to the correspondence between the user address text information and the location information, at least one location information corresponding to the same user address text information can be determined from the obtained multiple location information, and then based on at least one location information corresponding to the same user address text information, the location identified by the user address text information can be determined. That is to say, for any target user address text information, the user addresses corresponding to the multiple location information are matched with the target user address text information. If the match is consistent, the location information is determined as the reference location, and the next location information is matched; if the match is inconsistent, the location information is skipped, and the next location information is matched. Thus, at least one reference location corresponding to the target user address text information can be determined from the multiple location information, and the user addresses corresponding to the determined reference locations all match the target user address text information. Therefore, the target location identified by the target user address text information can be located according to the multiple reference locations.

[0105] Optionally, the target user address text information may be a POI place name, that is, it belongs to the target area. Or, optionally, the target user address text information may be a place name including a POI name and a building number, that is, it belongs to a specific target location. This embodiment does not limit this either. It should be noted that when the scope to which the target user address text information belongs is different, that is, the target user address text information belongs to the target area, or the target user address text information belongs to a specific target location, there will be a certain difference in the number of reference positions determined from multiple positioning positions, and thus the target position identified by the target user address text information obtained by positioning will also be different. Specifically, when the target user address text information belongs to the target area, the target position identified by the target user address text information may be the range of the identified area; when the target user address text information belongs to a specific target location, the target position identified by the target user address text information may be the positions of multiple objects within the identified area.

[0106] It can be understood that since the reference position of the target user address text information describes the possible positions identified by the target user address text information, the target position identified by the target user address text information can be located according to the reference position of the target user address text information. Since the reference position of the target user address text information can be one or more, when the reference position of the target user address text information is one, this reference position is the target position identified by the target user address text information; when the reference position of the target user address text information is multiple, as a possible implementation, the center of gravity of the multiple reference positions can be determined as the target position identified by the target user address text information.

[0107] The location determination method provided in this embodiment realizes determining the correspondence between the user address text information and the positioning position information according to the login information, the user address text information, and the positioning position information by obtaining the user address text information associated with multiple login information and the positioning position information of the clients logged in by the multiple login information. Thus, according to the correspondence between the user address text information and the positioning position information, the position identified by the user address text information is determined. Since the user address text information is associated with the login information, and the positioning position information is also the positioning position information of the client logged in by the login information, it is possible to determine the position identified by the user address text information without manual collection, solving the technical problem of needing to collect position information manually in the related art.

[0108] As can be seen from the above analysis, the target location identified by the user address text information may include the range of the identified area, and may also include the locations of multiple objects within the identified area. To clearly illustrate how to determine the range of the identified area based on the correspondence between the user address text information and the positioning location information when the target location identified by the user address text information includes the range of the identified area, this embodiment provides another location determination method. Figure 4 FIG. is a schematic flowchart of another location determination method provided by an embodiment of the present disclosure. In this embodiment, the location identified by the user address text information includes the range of the identified area.

[0109] As Figure 4 shown, the location determination method may include the following steps:

[0110] Step 401, obtain the user address text information associated with multiple login information.

[0111] Step 402, obtain the positioning location information of the clients logged in with multiple login information.

[0112] Step 403, determine the correspondence between the user address text information and the positioning location information according to the login information, the user address text information, and the positioning location information.

[0113] It should be noted that the execution processes of steps 401-403 can refer to the execution processes of embodiments 201-203, with the same principle and will not be elaborated here.

[0114] Step 404, according to the correspondence, determine the positioning location information that matches the name of the identified area from multiple positioning location information.

[0115] In this embodiment, the location identified by the user address text information includes the range of the identified area, so that the positioning location information that matches the name of the identified area can be determined from multiple positioning location information according to the correspondence between the user address text information and the positioning location information. Optionally, the positioning location information that matches the name of the identified area can be determined from multiple positioning location information according to the user address text information corresponding to the multiple positioning location information. As a possible implementation, the user address text information corresponding to the multiple positioning location information can be matched with the name of the identified area. If the match is consistent, the positioning location information is determined as the positioning location information that matches the name of the identified area, and the next positioning location information is matched; if the match is inconsistent, the positioning location information is skipped and the next positioning location information is matched. Thus, the positioning location information that matches the name of the identified area can be determined from multiple positioning location information.

[0116] Step 405: Set a selection box according to the positioning position information that matches the name of the identification area.

[0117] It can be understood that since the size of the POI is usually relatively stable and the shape is relatively regular, a selection box with a fixed size can be set based on the positioning position information that matches the name of the identification area determined in the previous step, so that as much positioning position information that matches the name of the identification area as possible is included in the selection box. As a possible implementation, a reference position <x m , y m > can be found from the positioning position information that matches the name of the identification area in the map, so that the rectangular box with the size of d l × d l , that is, the rectangle [<x m , y m >, <x m + d l , y m + d l >, can maximize the number of positioning position information that matches the name of the identification area it can cover. Among them, d l can be selected as a fixed value within the range of 600m to 1000m.

[0118] Step 406: Adjust the position of the selection box until the number of positioning position information within the selection box is maximized.

[0119] Since it is difficult to immediately find the rectangular box where <x m , y m > is located to maximize the number of positioning position information that matches the name of the identification area it can cover, it is necessary to continuously adjust the position of the selection box until the number of positioning position information within the selection box is maximized.

[0120] To find the rectangular box where <x m , y m > is located to maximize the number of positioning position information it can cover, optionally, a brute-force enumeration method can be used, or, optionally, a segment tree method can be used. This embodiment does not limit this.

[0121] Step 407: Determine the range of the identification area according to the coverage range of the selection box when the number of positioning position information within the selection box is maximized.

[0122] In this embodiment, when the number of positioning position information within the selection box is maximized, the range of the identification area can be determined based on the coverage range of the selection box. As a possible implementation, the selection box can be used as the identification area, and the coverage range of the selection box can be used as the range of the identification area.

[0123] It can be understood that since the position identified by the user address text information includes the range of the identified area, and based on the correspondence between the user address text information and the positioning position information, among multiple positioning position information, the positioning position information that matches the name of the identified area is determined, and based on the positioning position information that matches the name of the identified area, the range of the identified area is determined. Therefore, when determining the range of the identified area, the position identified by the user address text information is also determined. That is to say, the range of the identified area is the position identified by the user address text information.

[0124] The position determination method provided in this embodiment determines, according to the correspondence, the positioning position information that matches the name of the identified area from multiple positioning position information, and thus sets a selection box based on the positioning position information that matches the name of the identified area. After adjusting the position of the selection box until the number of positioning position information within the selection box is maximized, the range of the identified area is determined according to the coverage range of the selection box when the number of positioning position information within the selection box is maximized. Thus, when the target position identified by the user address text information includes the range of the identified area, based on the correspondence between the user address text information and the positioning position information, the position identified by the user address text information can be determined.

[0125] For the sake of clear illustration Figure 4 In the shown embodiment, for the process of adjusting the position of the selection box in step 405 until the number of positioning position information within the selection box is maximized, this embodiment provides Figure 5 The flow chart of determining the position of the selection box by the brute-force enumeration method as shown in Figure 5 As shown, using the brute-force enumeration method to determine the position of the selection box may include the following steps:

[0126] Step 501, initialize <x, y> as the minimum longitude and latitude of the target city.

[0127] It should be noted that the target city can be the city indicated by the user address text information. To reduce the spatial overhead of brute-force enumeration, if there are districts under the target city, the target city can be narrowed down to the corresponding district. If there are counties under the target city and there are townships or towns under the counties, the target city can be narrowed down to the corresponding township or town.

[0128] Initializing <x, y> as the minimum longitude and latitude of the target city means setting <x, y> as the minimum longitude and latitude of the target city. Among them, x can be set as the longitude of the target city, and y can be set as the latitude of the target city. Thus, according to the candidate position <x, y>, a rectangular box with a size of d l ×d l can be obtained, that is, rectangle [<x, y>, <x + dl , y + d l >], and this rectangular box is a region selection box with a set fixed size. Among them, d l can select 600m to obtain a fixed value within the range of 1000m.

[0129] Step 502, translate <x, y> with a step size of △l to scan the entire city. Formally, <x, y> is iteratively updated according to the following formula:

[0130]

[0131] Translate <x, y> with a step size of △l to scan the entire city, and multiple rectangular boxes with a size of d l ×d l can be obtained with <x, y> as the vertex, that is, multiple selection boxes can be obtained. Among them, the scanning method can be: when x is less than x max , y remains unchanged, and x is successively added with the step size △l to obtain the corresponding <x, y>, thereby obtaining the corresponding rectangular box; when x is greater than x max , x is initialized to the minimum value and remains unchanged, and y is successively added with the step size △l to obtain the corresponding <x, y>, thereby obtaining the corresponding rectangular box. Among them, x max is the maximum longitude value of the target city, and x min is the minimum longitude value of the target city.

[0132] It should be understood that the number of positioning position information covered by the multiple obtained selection boxes is different, so a selection box that can maximize the number of covered positioning position information can be determined from the multiple obtained selection boxes.

[0133] Step 503, while translating <x, y>, calculate the number of positioning position information covered by the corresponding selection box. If a new maximum number of positioning position information appears, use the current <x, y> as the return result.

[0134] Here, first calculate the number of positioning position information covered by the rectangular box corresponding to the initialized <x, y>, that is, the selection box, and set it as the maximum number of positioning position information. While translating <x, y>, calculate the number of positioning position information covered by the corresponding selection box. If the number of positioning position information covered by the selection box is greater than the maximum number of positioning position information, replace the maximum number of positioning position information with the number of positioning position information covered by this selection box, and use the current <x, y> as the return result for recording. Thus, after scanning the entire city, a selection box that can maximize the number of covered positioning position information is obtained.

[0135] In this embodiment, by initializing <x,y> to the minimum longitude and latitude values of the target city, <x,y> is translated by a step size of △l to scan the entire city, thereby obtaining multiple region selection frames with <x,y> as the vertex and a size of d l ×d l While translating <x,y>, calculate the number of location position information covered by the corresponding selection frame, and then determine the position of the selection frame according to the size of the number of location position information covered by the selection frame. Since the entire city is scanned at a fixed step size, a large number of states are examined, and even all states are exhausted, so that the determined selection frame can maximize the number of location position information covered.

[0136] The above embodiment describes how to determine the position identified by the user address text information based on the correspondence between the user address text information and the location position information when the target position identified by the user address text information includes the range of the identified area. To clearly illustrate how to determine the position identified by the user address text information based on the correspondence between the user address text information and the location position information when the target position identified by the user address text information also includes the positions of multiple objects within the identified area, this embodiment provides another position determination method. Figure 6 It is a schematic flowchart of another position determination method provided by the embodiments of the present disclosure. In this embodiment, the position identified by the user address text information also includes the positions of multiple objects within the identified area.

[0137] As Figure 6 shown, the position determination method may include the following steps:

[0138] Step 601, obtain the user address text information associated with multiple login information.

[0139] Step 602, obtain the location position information of the clients logged in with multiple login information.

[0140] Step 603, determine the correspondence between the user address text information and the location position information according to the login information, the user address text information, and the location position information.

[0141] Step 604, according to the correspondence, determine the location position information that matches the name of the identified area from multiple location position information.

[0142] It should be noted that the execution processes of steps 601-604 can refer to the execution processes of embodiments 601-604, with the same principle and will not be elaborated here.

[0143] Step 605, determine the range of the identified area according to the location position information that matches the name of the identified area.

[0144] In this embodiment, the range of the identification area can be determined based on the positioning location information that matches the name of the identification area. Specifically, a selection box that can contain as much positioning location information as possible that matches the name of the identification area can be set according to the positioning location information that matches the name of the identification area, and by continuously adjusting the position of the selection box, the number of positioning location information within the selection box can be maximized. Then, when the number of positioning location information within the selection box is maximized, the range of the identification area is determined based on the coverage of the selection box.

[0145] Step 606: For any one object, determine the position of the corresponding object according to the positioning location information that matches the name of the object.

[0146] Among them, the object can be understood as an object within the identification area, and the number is multiple.

[0147] In this embodiment, for any one object, the position of the corresponding object can be determined according to the positioning location information that matches the name of the object.

[0148] As a possible implementation, for any one object, according to the range of the identification area determined in the previous step, the user address text information corresponding to the positioning location information within the range of the identification area is matched with the name of the object. Similar to the matching method in step 404 of the previous embodiment, if the match is consistent, the positioning location information is determined as the position of the object, and the next positioning location information is matched; if the match is inconsistent, the positioning location information is skipped, and the next positioning location information is matched, so as to determine the position of the corresponding object according to the positioning location information that matches the name of the object. It should be noted that the positioning location information determined by this matching method that matches the name of the object can be one, or can be multiple, and this embodiment does not limit this.

[0149] Since the positioning location information determined to match the name of the object can be one, or can be multiple, when the positioning location information determined to match the name of the object is one, this positioning location information is the position of the object; when the positioning location information determined to match the name of the object is multiple, as a possible implementation, the centroid position of the multiple positioning location information can be determined as the position of the object. For example, assume that the positioning location information determined to match the name of the object is <x i , y i >, where the value of i is 1, 2, 3,..., then the centroid position <x bar , y bar > of the multiple positioning location information can be calculated to determine the position of the object. Among them, calculating the centroid position <x bar, y bar The formula of > is as follows:

[0150]

[0151] Step 607: Use a recurrent neural network to correct the positions of multiple objects.

[0152] When there are multiple objects in the identification area, the method of calculating the centroid to determine the positions of the objects will be affected by outliers or GPS errors, resulting in certain errors in the determined positions of the objects. Therefore, it is necessary to use a recurrent neural network to correct the positions of the objects. For a recurrent neural network, its input can be an indefinite-length sequence, but its output is of a fixed length, thus playing a role in correcting the positions of the objects.

[0153] The position determination method provided in this embodiment determines the positioning position information that matches the name of the identification area from multiple positioning position information according to the corresponding relationship, and thus determines the range of the identification area according to the positioning position information that matches the name of the identification area. After determining the positions of the corresponding objects according to the positioning position information that matches the name of the object for any one object, a recurrent neural network is used to correct the positions of multiple objects. Thus, when the target position identified by the user address text information also includes the positions of multiple objects in the identification area, the position identified by the user address text information can be determined based on the corresponding relationship between the user address text information and the positioning position information.

[0154] For the sake of clear illustration Figure 6 In the illustrated embodiment, the correction process of the recurrent neural network in step 607 is provided in this embodiment Figure 7 The schematic diagram of the correction process of the recurrent neural network shown in Figure 7 As shown, using a recurrent neural network to correct the positions of objects may include the following steps:

[0155] Step 701: Generate a position sequence according to the positions of multiple objects.

[0156] Here, there are multiple objects in the identification area, so there are the positions of multiple objects. Optionally, as a possible implementation, the positions of multiple objects can be sorted according to a certain rule to generate a position sequence.

[0157] Step 702: Use a recurrent neural network to perform position correction on the position sequence to obtain the output corrected position sequence.

[0158] Here, a recurrent neural network can be used to correct the position sequence generated based on the positions of multiple objects to obtain the output corrected position sequence. As a possible implementation, each position in the position sequence can be input into the recurrent neural network of the encoder for encoding to obtain the hidden state corresponding to each target position in the position sequence, so that the hidden state corresponding to each position in the position sequence is input into the recurrent neural network of the decoder for decoding to obtain the corrected data corresponding to the corrected position sequence.

[0159] It should be noted that the encoder is several layers of RNN (Recurrent Neural Network), which is used to process sequence data. Among them, the network parameters of the encoder are obtained by optimizing the loss function with the real annotation data. Similarly, the decoder is also several layers of RNN, and the network parameters of the decoder are also obtained by optimizing the loss function with the real annotation data.

[0160] Step 703, correct the corresponding positions in the position sequence according to the corrected data in the corrected position sequence.

[0161] Optionally, the corresponding positions in the position sequence can be corrected according to the corrected data in the corrected position sequence. As a possible implementation, the corrected data corresponding to each position in the position sequence can be added to each position data to obtain the corrected result as the final positions of each, thus playing a role in correcting the corresponding positions in the position sequence.

[0162] In this embodiment, by generating a position sequence according to the positions of multiple objects, each position in the position sequence is input into the recurrent neural network of the encoder for encoding to obtain the hidden state corresponding to each position in the position sequence, so that the hidden state corresponding to each position in the position sequence is input into the recurrent neural network of the decoder for decoding to obtain the corrected data corresponding to each position in the position sequence, and then the corresponding positions in the position sequence are corrected according to the corrected data in the corrected position sequence. Since in the case where there are multiple objects in the identification area, the positions of the multiple objects are usually regular, the recurrent neural network can be used to learn its rules to obtain the corrected data, and then the positions of the objects are corrected according to the corrected data.

[0163] To illustrate the above embodiments more clearly, an example is given below.

[0164] Such as Figure 8As shown in the figure, taking xx Community as an example, this location determination method can be divided into three parts. First, based on the positioning location information of the clients logged in with multiple login information and the user address text information associated with the multiple login information, obtain the user positioning locations belonging to all building numbers and place names under xx Community, that is, the positioning location information corresponding to each user address text information of "Building x in xx Community". Second, based on the distribution of the user positioning locations, find a <x m ,y m > rectangular box from multiple positioning location information, so that the number of positioning location information that can be covered by the rectangular box is maximized. Then, according to the coverage range of the rectangular box, determine the range of the identification area and filter out the positioning location information far from the community. Third, for the corresponding positioning location of each building, calculate its centroid and use the centroid of each building as the geographical location of the building. For example, for Building 1, find the centroid of its corresponding positioning location to get C1 in the figure, and use C1 as the geographical location of Building 1. The geographical locations of Buildings 2, 3, and 4 are obtained in the same way, getting C2, C3, and C4 in the figure.

[0165] Since the method of using the centroid to estimate the building coordinates will be affected by outliers or GPS errors. For example Figure 4 C3 in the figure is slightly north of the actual location of Building 3, so it is necessary to use a recurrent neural network model to learn how to fine-tune the building coordinates to make the result more accurate.

[0166] As Figure 9 shown, for each building number <c1, c2, …, c n > of each POI, it can be encoded through the recurrent neural network of the encoder to obtain the corresponding hidden state h. Then, input the hidden state h into the recurrent neural network of the decoder for decoding to obtain the corresponding correction data <s1, s2, …, s n >. Add the correction data to the original coordinates to obtain the corrected result and use it as the final coordinate position. Among them, the network parameters of the encoder and decoder are obtained by optimizing the loss function with the real annotation data.

[0167] In this embodiment, based on the positioning location of the clients logged in with multiple login information and the associated user address text, obtain the corresponding user positioning locations belonging to all building numbers and place names under the community, realize determining the reference location of the community from multiple positioning locations, and then calculate the centroid according to the corresponding user positioning location of each building to locate its geographical location, obtain the rough positioning coordinates of the building, and further correct the rough positioning coordinates of the building through a recurrent neural network. Since the building number rules can be learned through the recurrent network to correct the rough positioning coordinates of the building, the influence of using the method of finding the centroid to estimate the building coordinates can be effectively reduced, and the positions of each building can be accurately located.

[0168] To implement the above embodiments, the present disclosure also proposes a position determination device.

[0169] Figure 10 The following is a schematic structural diagram of a position determination device provided by an embodiment of the present disclosure.

[0170] As Figure 10 shown, the position determination device includes: a first acquisition module 11, a second acquisition module 12, a determination module 13, and a positioning module 14.

[0171] The first acquisition module 11 is configured to acquire user address text information associated with multiple login information.

[0172] The second acquisition module 12 is configured to acquire positioning location information of the client logged in with multiple login information.

[0173] The determination module 13 is configured to determine the correspondence between the user address text information and the positioning location information according to the login information, the user address text information, and the positioning location information.

[0174] The positioning module 14 is configured to determine the location identified by the user address text information according to the correspondence between the user address text information and the positioning location information.

[0175] Further, in a possible implementation manner of the embodiment of the present disclosure, the location identified by the user address text information includes the range of the identified area; the positioning module 14 includes: a first determination unit 1401 and a second determination unit 1402.

[0176] The first determination unit 1401 is configured to determine, according to the correspondence, the positioning location information that matches the name of the identified area from multiple positioning location information.

[0177] The second determination unit 1402 is configured to determine the range of the identified area according to the positioning location information that matches the name of the identified area.

[0178] Further, in a possible implementation manner of the embodiment of the present disclosure, the second determination unit 1402 is configured to:

[0179] Adjust the position of the selection box until the number of positioning location information within the selection box is maximized;

[0180] Determine the range of the identified area according to the coverage range of the selection box when the number of positioning location information within the selection box is maximized.

[0181] It should be noted that the foregoing explanation of the embodiment of the position determination method is also applicable to the position determination device of this embodiment, and will not be repeated here.

[0182] Based on the above embodiments, the embodiments of the present disclosure also provide a possible implementation of a position determination device. Figure 11 FIG. is a schematic structural diagram of another position determination device provided by the embodiments of the present disclosure. On the basis of the previous embodiment, the position determination device further includes: a correction module 15.

[0183] The correction module 15 includes: a third determination unit 1501, a generation unit 1502, a first correction unit 1503, and a second correction unit 1504.

[0184] The third determination unit 1501 is configured to determine the position of a corresponding object according to the positioning position information matching the name of the object for any one object.

[0185] The generation unit 1502 is configured to generate a position sequence according to the positions of multiple objects.

[0186] The first correction unit 1503 is configured to perform position correction on the position sequence by using a recurrent neural network to obtain an output corrected position sequence.

[0187] The second correction unit 1504 is configured to correct the corresponding positions in the position sequence according to the correction data in the corrected position sequence.

[0188] Further, in a possible implementation manner of the embodiments of the present disclosure, the first correction unit 1503 is configured to:

[0189] Input each position in the position sequence into the recurrent neural network of the encoder for encoding to obtain the hidden state corresponding to each position in the position sequence.

[0190] Input the hidden state corresponding to each position in the position sequence into the recurrent neural network of the decoder for decoding to obtain the corresponding correction data in the corrected position sequence.

[0191] Further, in a possible implementation manner of the embodiments of the present disclosure, the third determination unit 1501 is configured to:

[0192] For any one object, if there are multiple pieces of positioning position information matching the name of the object, determine the position of the object according to the centroid position of the multiple pieces of positioning position information.

[0193] The location determination device provided by the embodiments of the present disclosure realizes determining the corresponding relationship between the user address text information and the positioning location information according to the login information, the user address text information, and the positioning location information by obtaining the user address text information associated with multiple login information and the positioning location information of the clients logged in by the multiple login information. Thus, according to the corresponding relationship between the user address text information and the positioning location information, the location identified by the user address text information is determined. Since the user address text information is associated with the login information, and the positioning location information obtained is also the positioning location information of the client logged in by the login information, it is possible to determine the location identified by the user address text information without manual collection, solving the technical problem in the related art that location information needs to be collected manually.

[0194] To implement the above embodiments, the present disclosure also proposes an electronic device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein, the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the location determination method proposed in any of the above embodiments of the present disclosure.

[0195] Figure 12 The structure diagram of an electronic device provided by the embodiments of the present disclosure can implement the process of the embodiments shown in the present disclosure Figure 1-11 as shown in Figure 12 As shown, the electronic device may include: a housing 1201, a processor 1202, a memory 1203, a circuit board 1204, and a power supply circuit 1205. Among them, the circuit board 1204 is arranged inside the space surrounded by the housing 1201, and the processor 1202 and the memory 1203 are arranged on the circuit board 1204; the power supply circuit 1205 is used to supply power to each circuit or device of the above electronic device; the memory 1203 is used to store executable program codes; the processor 1202 runs the program corresponding to the executable program code by reading the executable program codes stored in the memory 1203, and is used to execute the location determination method described in any of the foregoing embodiments.

[0196] The specific execution process of the above steps by the processor 1202 and the further steps executed by the processor 1202 by running the executable program code can refer to the description of the embodiments shown in the present disclosure Figure 1-9 and will not be elaborated herein.

[0197] To implement the above embodiments, the present disclosure also proposes a computer-readable storage medium storing computer instructions, wherein the computer instructions are used to cause the computer to execute the location determination method proposed in any of the above embodiments of the present disclosure.

[0198] To implement the above embodiments, the present disclosure also provides a computer program product, including a computer program which, when executed by a processor, implements the location determination method provided in any one of the above embodiments of the present disclosure.

[0199] In the description of this specification, the descriptions referring to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples", etc. mean that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present disclosure. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, without conflict, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of different embodiments or examples.

[0200] In addition, the terms "first" and "second" are used for descriptive purposes only and cannot be construed as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include at least one of such features. In the description of the present disclosure, "a plurality" means at least two, such as two, three, etc., unless otherwise specifically and clearly defined.

[0201] Any process or method description in a flowchart or described in other ways herein may be understood to represent a module, segment, or portion of code including one or more executable instructions for implementing a customized logical function or process, and the scope of the preferred embodiments of the present disclosure includes additional implementations, where the functions may be executed in a manner substantially simultaneous with or in a reverse order according to the involved functions, rather than in the order shown or discussed, which should be understood by those skilled in the art to which the embodiments of the present disclosure pertain.

[0202] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a definable sequence list of executable instructions for implementing logical functions, and can be specifically implemented in any computer-readable medium for use by an instruction execution system, apparatus, or device (such as a computer-based system, a system including a processor, or other systems that can fetch and execute instructions from the instruction execution system, apparatus, or device), or used in combination with these instruction execution systems, apparatus, or devices. For the purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. More specific examples (non-exhaustive list) of computer-readable media include the following: an electrical connection portion with one or more wirings (electronic device), a portable computer diskette (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable medium can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other media, then editing, interpreting, or otherwise processing it as appropriate, and then storing it in a computer memory.

[0203] It should be understood that various parts of the present disclosure can be implemented by hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, any one or a combination of the following techniques well known in the art can be used: discrete logic circuits having logic gate circuits for implementing logical functions on data signals, application specific integrated circuits having appropriate combinational logic gate circuits, programmable gate arrays (PGAs), field programmable gate arrays (FPGAs), etc.

[0204] Those of ordinary skill in the art of this technology can understand that all or part of the steps carried by the methods of the above embodiments can be completed by instructing relevant hardware through a program, and the program can be stored in a computer-readable storage medium. When the program is executed, it includes one or a combination of the steps of the method embodiments.

[0205] In addition, each functional unit in various embodiments of the present disclosure may be integrated into one processing module, may exist physically alone for each unit, or two or more units may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. When the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0206] The above-mentioned storage medium may be a read-only memory, a magnetic disk, an optical disc, etc. Although the embodiments of the present disclosure have been shown and described above, it can be understood that the above embodiments are exemplary and should not be construed as limiting the present disclosure. Those of ordinary skill in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present disclosure.

Claims

1. A position determination method, characterized in that, Including the following steps: Obtain user address text information associated with multiple login information, where the locations identified by the user address text information include the scope of the identification area and the locations of multiple objects within the identification area; Obtain the positioning location information of the clients logged in with multiple pieces of the login information; Determine the correspondence between the user address text information and the positioning location information according to the login information, the user address text information, and the positioning location information; Determine the location identified by the user address text information according to the correspondence between the user address text information and the positioning location information; Among them, determining the location identified by the user address text information according to the correspondence between the user address text information and the positioning location information includes: According to the correspondence, determine the positioning location information that matches the name of the identification area from multiple pieces of the positioning location information; Determine the scope of the identification area according to the positioning location information that matches the name of the identification area; For any one object, determine the location of the corresponding object according to the positioning location information within the scope of the identification area that matches the name of the object; Generate a location sequence according to the locations of multiple objects; Use a recurrent neural network to correct the locations of the location sequence to obtain an output corrected location sequence; Correct the corresponding locations in the location sequence according to the correction data in the corrected location sequence.

2. The method according to claim 1, characterized in that, The determining the scope of the identification area according to the positioning location information that matches the name of the identification area includes: Adjust the position of the selection box until the number of positioning location information within the selection box is maximized; Determine the scope of the identification area according to the coverage of the selection box when the number of positioning location information within the selection box is maximized.

3. The method according to claim 1, characterized in that, The using a recurrent neural network to correct the locations of the location sequence to obtain an output corrected location sequence includes: Input the locations in the location sequence into the recurrent neural network of the encoder for encoding to obtain the hidden states corresponding to the locations in the location sequence; Input the hidden states corresponding to the locations in the location sequence into the recurrent neural network of the decoder for decoding to obtain the correction data corresponding to the locations in the corrected location sequence.

4. The method according to claim 1, characterized in that, The for any one object, determining the location of the corresponding object according to the positioning location information that matches the name of the object includes: For any one object, if there are multiple pieces of positioning location information that match the name of the object, determine the location of the object according to the centroid location of the multiple pieces of positioning location information.

5. A position determination device, characterized in that, Including: A first acquisition module for obtaining user address text information associated with multiple login information, where the locations identified by the user address text information include the scope of the identification area and the locations of multiple objects within the identification area; A second acquisition module for obtaining the positioning location information of the clients logged in with multiple pieces of the login information; A determination module, configured to determine a correspondence between the user address text information and the positioning location information according to the login information, the user address text information, and the positioning location information; A positioning module, configured to determine a location identified by the user address text information according to the correspondence between the user address text information and the positioning location information; The positioning module includes: A first determination unit, configured to determine, according to the correspondence, positioning location information that matches the name of the identification area from multiple pieces of the positioning location information; A second determination unit, configured to determine a range of the identification area according to the positioning location information that matches the name of the identification area; The apparatus further includes a correction module, including: A third determination unit, configured to determine a location of a corresponding object according to the positioning location information that matches the name of the object for any one object; A generation unit, configured to generate a location sequence according to the locations of multiple objects; A first correction unit, configured to perform location correction on the location sequence by using a recurrent neural network to obtain an output corrected location sequence; A second correction unit, configured to correct a corresponding location in the location sequence according to correction data in the corrected location sequence.

6. The device according to claim 5, characterized in that, The second determination unit is configured to: Adjust the position of the selection box until the number of positioning location information within the selection box is maximized; Determine the range of the identification area according to the coverage range of the selection box when the number of positioning location information within the selection box is maximized.

7. The device according to claim 5, wherein, The first correction unit is configured to: Input each location in the location sequence into a recurrent neural network of an encoder for encoding to obtain hidden states corresponding to each location in the location sequence; Input the hidden states corresponding to each location in the location sequence into a recurrent neural network of a decoder for decoding to obtain correction data corresponding to the corrected location sequence.

8. The device according to any one of claims 5 - 7, wherein, The third determination unit is configured to: For any one object, if there are multiple pieces of positioning location information that match the name of the object, determine the location of the object according to the centroid location of the multiple pieces of positioning location information.

9. An electronic device, comprising: At least one processor; And A memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor executes the method according to any one of claims 1-4.

10. A non - transitory computer - readable storage medium storing computer instructions, wherein, The computer instructions are used to cause the computer to execute the method according to any one of claims 1-4.

11. A computer program product, comprising a computer program, wherein the computer program, when executed by a processor, implements the method according to any one of claims 1 - 4.

Images