Method and device for identifying multiple accounts of one person

Abstract

The present application discloses a method and device for identifying multiple accounts for one person. A specific implementation of the method includes: obtaining user information corresponding to the accounts in an account pair and the identification information involved; determining whether the accounts in the account pair satisfy preset association information, the preset association information is used to characterize whether the accounts in the account pair have the possibility of belonging to the same person; in response to determining that the accounts in the account pair satisfy the preset association information, determining whether the accounts in the account pair satisfy preset judgment conditions, the preset judgment conditions are used to determine whether the accounts in the account pair belong to the same person; in response to determining that the accounts in the account pair do not satisfy the preset judgment conditions, processing the user information and identification information of the accounts in the account pair to obtain a feature vector corresponding to the account pair; inputting the feature vector into a recognition model to determine whether the accounts in the account pair belong to the same person. The present application provides a method for identifying multiple accounts for one person, which improves the recognition accuracy.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of computer technology, and more particularly to a method and apparatus for identifying multiple accounts for one person. Background Art

[0002] It is very common for the same person to have multiple accounts on e-commerce platforms. For example, in addition to a personal account, a user on JD.com also has a corporate account for conducting business. Another example is that a user has registered multiple personal accounts on different devices. Summary of the invention

[0003] The embodiments of the present application provide a method and device for identifying multiple accounts of one person.

[0004] In a first aspect, an embodiment of the present application provides a method for identifying multiple accounts for one person, including: obtaining user information and related identification information corresponding to the accounts in an account pair; determining whether the accounts in the account pair satisfy preset association information, the preset association information being used to characterize whether the accounts in the account pair have the possibility of belonging to the same person; in response to determining that the accounts in the account pair satisfy the preset association information, determining whether the accounts in the account pair satisfy preset judgment conditions, wherein the preset judgment conditions are used to determine whether the accounts in the account pair belong to the same person; in response to determining that the accounts in the account pair do not satisfy the preset judgment conditions, processing the user information and identification information of the accounts in the account pair to obtain a feature vector corresponding to the account pair; inputting the feature vector into a pre-trained recognition model to determine whether the accounts in the account pair belong to the same person, wherein the recognition model is used to characterize the correspondence between the feature vector corresponding to the account pair and the judgment result of whether the accounts in the account pair belong to the same person.

[0005] In some embodiments, the preset determination condition includes a first determination condition and a second determination condition, the first determination condition is used to determine that the accounts in the account pair belong to the same person, and the second determination condition is used to determine that the accounts in the account pair do not belong to the same person.

[0006] In some embodiments, the above-mentioned processing of user information and identification information of the accounts in the account pair to obtain a feature vector corresponding to the account pair includes: processing corresponding item information between the accounts in the account pair to obtain a sub-feature vector corresponding to each corresponding item information; splicing each sub-feature vector to obtain a feature vector.

[0007] In some embodiments, user information includes user portrait information, consumption habit information and receipt information; the above-mentioned processing of corresponding item information between accounts in the account pair obtains a sub-feature vector corresponding to each corresponding item information, including: determining a sub-feature vector representing the similarity of corresponding item information between accounts based on the user information of the accounts in the account pair.

[0008] In some embodiments, the above-mentioned processing of corresponding item information between accounts in the account pair to obtain a sub-feature vector corresponding to each corresponding item information includes: for each type of identification information, performing the following operations: determining a sub-feature vector representing the number of intersection identification information involved in all accounts in the account pair based on the identification information of the type of the account in the account pair; for each account in the account pair, determining a sub-feature vector representing the degree of belonging of the intersection identification information relative to the account based on the number of associations between the account and the intersection identification information and the number of associations with all identification information of the type; determining a sub-feature vector representing the sum of the degrees of belonging of the accounts in the account pair and a sub-feature vector representing the difference in the degrees of belonging of the accounts in the account pair; for each identification information of the type, determining a sub-feature vector representing the degree of sharing of the identification information based on the number of accounts associated with the identification information.

[0009] In some embodiments, the recognition model is trained in the following manner: obtaining user information and identification information corresponding to each account in an account set; combining accounts in the account set that meet preset association information into account pairs to obtain multiple account pairs; screening account pairs that meet preset judgment conditions from the multiple account pairs to obtain multiple training account pairs, and setting a label for each training account pair according to the preset judgment conditions to represent whether the accounts in the training account pair belong to the same person; processing the user information and identification information of the account in each training account pair in the multiple account pairs to obtain a feature vector corresponding to the training account pair; using a machine learning method, with the feature vector corresponding to the training account pair as input and the label corresponding to the input training account pair as the expected output, to train the initial recognition model to obtain a recognition model.

[0010] In a second aspect, an embodiment of the present application provides a device for identifying multiple accounts of one person, comprising: an acquisition unit, configured to acquire user information and identification information corresponding to the accounts in an account pair; a first determination unit, configured to determine whether the accounts in the account pair satisfy preset association information, the preset association information being used to characterize whether the accounts in the account pair have the possibility of belonging to the same person; a second determination unit, configured to determine whether the accounts in the account pair satisfy preset judgment conditions in response to determining that the accounts in the account pair satisfy the preset association information, wherein the preset judgment conditions are used to determine whether the accounts in the account pair belong to the same person; a processing unit, configured to process the user information and identification information of the accounts in the account pair in response to determining that the accounts in the account pair do not satisfy the preset judgment conditions, and obtain a feature vector corresponding to the account pair; a third determination unit, configured to input the feature vector into a pre-trained recognition model to determine whether the accounts in the account pair belong to the same person, wherein the recognition model is used to characterize the correspondence between the feature vector corresponding to the account pair and the judgment result of whether the accounts in the account pair belong to the same person.

[0011] In some embodiments, the preset determination condition includes a first determination condition and a second determination condition, the first determination condition is used to determine that the accounts in the account pair belong to the same person, and the second determination condition is used to determine that the accounts in the account pair do not belong to the same person.

[0012] In some embodiments, the processing unit is further configured to: process corresponding item information between accounts in the account pair to obtain a sub-feature vector corresponding to each corresponding item information; and concatenate each sub-feature vector to obtain a feature vector.

[0013] In some embodiments, the user information includes user portrait information, consumption habit information and receipt information; the processing unit is further configured to: determine a sub-feature vector representing the similarity of corresponding item information between the accounts based on the user information of the accounts in the account pair.

[0014] In some embodiments, the processing unit is further configured to: perform the following operations for each type of identification information: determine a sub-feature vector representing the number of intersection identification information involved in all accounts in the account pair based on the identification information of the type of the account in the account pair; for each account in the account pair, determine a sub-feature vector representing the degree of belonging of the intersection identification information relative to the account based on the number of associations between the account and the intersection identification information and the number of associations with all identification information of the type; determine a sub-feature vector representing the sum of the degrees of belonging of the accounts in the account pair and a sub-feature vector representing the difference in the degrees of belonging of the accounts in the account pair; for each identification information of the type, determine a sub-feature vector representing the degree of sharing of the identification information based on the number of accounts associated with the identification information.

[0015] In some embodiments, the above-mentioned device also includes: a training unit, which is configured to train and obtain a recognition model in the following manner: obtain user information and related identification information corresponding to each account in the account set; combine accounts in the account set that meet preset association information into account pairs to obtain multiple account pairs; screen out account pairs that meet preset judgment conditions from multiple account pairs to obtain multiple training account pairs, and set a label for each training account pair according to the preset judgment conditions to represent whether the accounts in the training account pair belong to the same person; process the user information and identification information of the account in each training account pair in the multiple account pairs to obtain a feature vector corresponding to the training account pair; use a machine learning method, with the feature vector corresponding to the training account pair as input and the label corresponding to the input training account pair as the expected output, to train the initial recognition model and obtain a recognition model.

[0016] In a third aspect, an embodiment of the present application provides a computer-readable medium having a computer program stored thereon, wherein when the program is executed by a processor, the method described in any implementation manner of the first aspect is implemented.

[0017] In a fourth aspect, an embodiment of the present application provides an electronic device, comprising: one or more processors; a storage device on which one or more programs are stored, and when the one or more programs are executed by one or more processors, the one or more processors implement the method described in any implementation manner of the first aspect.

[0018] The method and device for identifying multiple accounts of one person provided in the embodiments of the present application obtain user information and identification information corresponding to the accounts in an account pair; determine whether the accounts in the account pair satisfy preset association information, the preset association information is used to characterize whether the accounts in the account pair have the possibility of belonging to the same person; in response to determining that the accounts in the account pair satisfy the preset association information, determine whether the accounts in the account pair satisfy preset judgment conditions, wherein the preset judgment conditions are used to determine whether the accounts in the account pair belong to the same person; in response to determining that the accounts in the account pair do not satisfy the preset judgment conditions, process the user information and identification information of the accounts in the account pair to obtain a feature vector corresponding to the account pair; input the feature vector into a pre-trained recognition model to determine whether the accounts in the account pair belong to the same person, wherein the recognition model is used to characterize the correspondence between the feature vector corresponding to the account pair and the judgment result of whether the accounts in the account pair belong to the same person, thereby providing a method for identifying multiple accounts of one person and improving recognition accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Other features, objects and advantages of the present application will become more apparent by reading the detailed description of non-limiting embodiments made with reference to the following drawings:

[0020] Figure 1 is an exemplary system architecture diagram in which an embodiment of the present application may be applied;

[0021] Figure 2 is a flow chart of an embodiment of a method for identifying multiple accounts of one person according to the present application;

[0022] Figure 3 is a schematic diagram of an application scenario of the method for identifying multiple accounts for one person according to this embodiment;

[0023] Figure 4 is a flow chart of another embodiment of a method for identifying multiple accounts of one person according to the present application;

[0024] Figure 5 is a structural diagram of an embodiment of a device for identifying multiple accounts of one person according to the present application;

[0025] Figure 6 It is a structural diagram of a computer system suitable for implementing the embodiments of the present application. DETAILED DESCRIPTION

[0026] The present application will be further described in detail below in conjunction with the accompanying drawings and embodiments. It is to be understood that the specific embodiments described herein are only used to explain the relevant invention, rather than to limit the invention. It should also be noted that, for ease of description, only the parts related to the relevant invention are shown in the accompanying drawings.

[0027] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0028] Figure 1 An exemplary architecture 100 is shown to which the method and apparatus for identifying multiple accounts of one person of the present application can be applied.

[0029] like Figure 1 As shown, the system architecture 100 may include terminal devices 101, 102, 103, a network 104 and a server 105. The communication connection between the terminal devices 101, 102, 103 constitutes a topological network, and the network 104 is used to provide a medium for the communication link between the terminal devices 101, 102, 103 and the server 105. The network 104 may include various connection types, such as wired, wireless communication links or optical fiber cables, etc.

[0030] Terminal devices 101, 102, 103 may be hardware devices or software that support network connection for data interaction and data processing. When terminal devices 101, 102, 103 are hardware, they may be various electronic devices that support network connection, information acquisition, interaction, display, processing and other functions, including but not limited to smart phones, tablet computers, e-book readers, laptop computers and desktop computers, etc. When terminal devices 101, 102, 103 are software, they may be installed in the electronic devices listed above. They may be implemented as multiple software or software modules for providing distributed services, or they may be implemented as a single software or software module. No specific limitation is made here.

[0031] Server 105 can be a server that provides various services, such as obtaining user information corresponding to the accounts associated on terminal devices 101, 102, and 103, the identification information involved, and determining a background processing server for multiple accounts for one person. The background processing server will need to determine whether two accounts belonging to the same person form an account pair, and determine whether the accounts in the account pair belong to the same person based on the triple recognition of whether the preset association information is met, whether the preset judgment conditions are met, and whether the recognition model is used for recognition. As an example, server 105 can be a cloud server.

[0032] It should be noted that the server can be hardware or software. When the server is hardware, it can be implemented as a distributed server cluster consisting of multiple servers, or it can be implemented as a single server. When the server is software, it can be implemented as multiple software or software modules (for example, software or software modules used to provide distributed services), or it can be implemented as a single software or software module. No specific limitation is made here.

[0033] It should also be noted that the method for identifying multiple accounts of one person provided by the embodiments of the present disclosure can be executed by a server, or by a terminal device, or by a server and a terminal device in cooperation with each other. Accordingly, the various parts (such as various units, sub-units, modules, and sub-modules) included in the apparatus for identifying multiple accounts of one person can be all set in the server, or all set in the terminal device, or can be set in the server and the terminal device respectively.

[0034] It should be understood that Figure 1 The number of terminal devices, networks, and servers in the system is only illustrative. Any number of terminal devices, networks, and servers may be provided as required. When the electronic device on which the method for identifying multiple accounts of one person is running does not need to transmit data with other electronic devices, the system architecture may only include the electronic device (e.g., server or terminal device) on which the method for identifying multiple accounts of one person is running.

[0035] Continue to refer Figure 2 , a process 200 of an embodiment of a method for identifying multiple accounts of one person is shown, comprising the following steps:

[0036] Step 201, obtaining the user information corresponding to the account in the account pair and the identification information involved.

[0037] In this embodiment, the execution subject of the method for identifying multiple accounts of one person (for example Figure 1 The server in the account pair can obtain the user information and the identification information involved, which correspond to the account in the account pair, remotely or locally, through a wired connection or a wireless connection.

[0038] The account may be any type of account, such as a JD account, a Taobao account, a QQ account, etc. The two accounts in the account pair are accounts that need to be determined whether they belong to the same person. For each account in the account pair, the execution subject may obtain the user information corresponding to the account and the identification information involved.

[0039] User information represents any information of the user that can be obtained based on the account. For example, user information includes registration information filled in by the user when registering an account. Identification information represents any type of hardware identification information and software identification information associated with the account. For example, identification information includes device ID (Identity document), OpenID (open identification number) and Union ID (union identification number) of mini programs, etc. Among them, device ID includes but is not limited to IMEI (International Mobile Equipment Identity) of Android system, AID (Andriod ID, identity identifier of Android system), IDFA (Identifier for Advertising) of IOS (mobile operating system developed by Apple), Open UDID (Unique Device Identifier, unique device identifier of the device), etc. In this embodiment, when the account has logged in to the hardware device corresponding to the hardware identification information, or logged in to the software corresponding to the software identification information, the identification information is considered to be the identification information involved or associated with the account.

[0040] As an example, the account is a JD account, and the above-mentioned execution entity can collect user information and identification information corresponding to the account from various business channels of the JD e-commerce platform.

[0041] Step 202: Determine whether the account in the account pair satisfies the preset association information.

[0042] In this embodiment, the execution subject may determine whether the accounts in the account pair satisfy the preset association information, wherein the preset association information is used to indicate whether the accounts in the account pair have the possibility of belonging to the same person.

[0043] The preset association information can be specifically set according to the account in the account pair. As an example, the preset association information can be that there is at least one identical identification information in the identification information involved in the two accounts. For example, the two accounts have logged in to the same device, or the two accounts have connected to the same wireless hotspot.

[0044] It can be understood that there is generally a correlation between two accounts belonging to the same person. The correlation between the accounts of the same person can be characterized by preset correlation information. When the two accounts in the account pair meet the preset correlation information, the above-mentioned execution subject can determine that there is a possibility that the accounts in the account pair belong to the same person; when the two accounts in the account pair do not meet the preset correlation information, the above-mentioned execution subject can determine that there is no possibility that the accounts in the account pair belong to the same person.

[0045] Step 203, in response to determining that the account in the account pair satisfies the preset association information, determining whether the account in the account pair satisfies the preset determination condition.

[0046] In this embodiment, the execution subject may determine whether the accounts in the account pair meet a preset determination condition in response to determining that the accounts in the account pair meet the preset association information. The preset determination condition is used to determine whether the accounts in the account pair belong to the same person.

[0047] The preset judgment condition can be specifically set according to the account in the account pair. As an example, the account is a user account of an e-commerce platform, and the preset judgment condition can be whether the registration information between the accounts is the same. Specifically, when the ID numbers in the registration information of the two accounts are different, it can be determined that the two accounts do not belong to the same person.

[0048] In some optional implementations of this embodiment, the preset determination condition includes a first determination condition and a second determination condition. Among them, the first determination condition is used to determine that the accounts in the account pair belong to the same person. When the accounts in the account pair meet the first determination condition, it can be determined that the accounts in the account pair belong to the same person; however, when the accounts in the account pair do not meet the first determination condition, it cannot be determined that the accounts in the account pair do not belong to the same person. The second determination condition is used to determine that the accounts in the account pair do not belong to the same person. When the accounts in the account pair meet the second determination condition, it can be determined that the accounts in the account pair do not belong to the same person; however, when the accounts in the account pair do not meet the second determination condition, it cannot be determined that the accounts in the account pair belong to the same person.

[0049] Continuing with the example of a user account on an e-commerce platform, the first judgment condition is that the phone number in the most frequent receiving information corresponding to the account is the same, and the gender and age of the user corresponding to the account are the same. When it is determined that the phone number in the most frequent receiving information corresponding to the account in the account pair is the same, and the gender and age of the user corresponding to the account are the same, then the accounts in the account pair are deemed to belong to the same person; however, when at least one of the phone number in the receiving information of the account in the account pair, the gender and age of the user corresponding to the account is different, it cannot be determined that the accounts in the account pair do not belong to the same person, and there is still a possibility that they belong to the same person. The second judgment condition is that the accounts in the account pair are the accounts corresponding to the members of the registered family accounts, or the ID numbers in the registration information of the accounts in the account pair are different. Among them, a family account is a plurality of accounts associated with a family relationship, and a plurality of the accounts have been authenticated as a family relationship.

[0050] When it is determined that the accounts in the account pair do not belong to the accounts corresponding to the members of the family account, or it cannot be determined that the ID numbers in the registration information of the accounts in the account pair are different, it cannot be determined that the accounts in the account pair belong to the same person.

[0051] It is understandable that the preset determination conditions may not be able to determine whether the accounts in the account pair belong to the same person. That is, the accounts in the account pair neither meet the first determination condition and are determined to belong to the same person, nor meet the second determination condition and are determined to not belong to the same person.

[0052] Step 204, in response to determining that the account in the account pair does not satisfy the preset determination condition, the user information and identification information of the account in the account pair are processed to obtain a feature vector corresponding to the account pair.

[0053] In this embodiment, in response to determining that an account in the account pair does not satisfy a preset determination condition, the execution subject may process the user information and identification information of the account in the account pair to obtain a feature vector corresponding to the account pair.

[0054] As an example, the above-mentioned execution entity may digitize the user information and identification information of the accounts in the account pair based on the same standard, and use the vector obtained after digitization as the feature vector corresponding to the account pair.

[0055] In some optional implementations of this embodiment, the execution subject performs step 204 in the following manner:

[0056] First, the corresponding item information between the accounts in the account pair is processed to obtain a sub-feature vector corresponding to each corresponding item information. The corresponding item information represents the corresponding information between the accounts in the account pair. For example, the account pair includes account x and account y, and the registration information corresponding to account x and the registration information corresponding to account y can be considered as corresponding item information. The above execution subject can digitize each set of corresponding item information to obtain a sub-feature vector corresponding to the corresponding item information.

[0057] Second, concatenate each sub-feature vector to obtain a feature vector.

[0058] As an example, the execution entity may concatenate each sub-feature vector based on a preset order to obtain a feature vector.

[0059] In some optional implementations of this embodiment, the user information corresponding to the account includes user portrait information, consumption habit information and receipt information. With respect to the above user information, the above execution subject may determine a sub-feature vector representing the similarity of corresponding item information between the accounts based on the user information of the accounts in the account pair.

[0060] It can be understood that when the accounts in an account pair belong to the same person, the similarity of the user information corresponding to the accounts in the account pair is higher; when the accounts in the account pair do not belong to the same person, the similarity of the user information corresponding to the accounts in the account pair is lower.

[0061] Taking the recipient information as an example, the above execution entity can count the proportion of orders with overlapping recipient information between the accounts in the account pair. For the recipient name, the number of orders for account number x is recorded as o x , the number of orders for account y is o y , the number of orders with the same recipient name for account number x and account number y is o xy The above execution entity obtains the proportion of overlapping recipient names in the orders of account number x, r x =o xy / o x , get the proportion of overlapping recipient names in the orders of account y y =o xy / o y The proportion of orders with overlapping recipient information depicts the possibility that the recipient information is the account holder's information. The higher the proportion, the more likely the recipient information is the account holder's information, and the higher the similarity between the two accounts.

[0062] For categorizable information in user information such as gender, purchasing power, and commonly used payment methods, the similarity between accounts in an account pair is distinguished by a specific value. Taking gender as an example, when the genders of account x and account y are both non-empty and the same, the similarity is determined to be 1; when the gender of account x and account y is empty, the similarity is determined to be 0; when the genders of account x and account y are both non-empty and different, the similarity is determined to be -1.

[0063] In some optional implementations of this embodiment, for each type of identification information, the execution subject may perform the following operations:

[0064] First, according to the identification information of the type of the account in the account pair, a sub-feature vector representing the amount of intersection identification information involved in the accounts in the account pair is determined.

[0065] As an example, the identification information of this type is the IMEI information of the Android system. The identification information of this type for account x in the account pair includes A, B, C, and D, and the identification information of this type for account y in the account pair includes B, C, D, and E. Then the intersection identification information of accounts x and y in the account pair is B, C, and D, and the number is 3. Then, the above execution entity can determine the sub-feature vector representing the number of intersection identification information involved in the accounts in the account pair

[0066] Second, for each account in the account pair, a sub-feature vector representing the degree of belonging of the intersection identification information to the account is determined based on the number of associations between the account and the intersection identification information and the number of associations with all identification information of this type.

[0067] In this implementation, each time an account logs into a hardware device corresponding to the hardware identification information or a software corresponding to the software identification information, it can be determined that the account is associated with the hardware device corresponding to the hardware identification information or the software corresponding to the software identification information once.

[0068] Continuing with the example of the IMEI information of the Android system, the IMEI set of devices logged in by account x is The IMEI set of devices logged in by account y is Then the set of IMEIs of the same device logged in by account x and account y is Where m and n are both positive integers. The total number of times account number x logs in on IMEI is but Among them, x i For account x in imei i The number of logins on the corresponding device. Similarly, the total number of logins of account y on IMEI is but Among them, yi For account y in imei i The total number of logins on the corresponding device. Determine the total number of logins for account x and account y on the same IMEI. and but We define the degree of attribution of IMEI to account x as Define the degree of attribution of IMEI to account y as Intuitively, for accounts x and y in the account pair, the degree of IMEI ownership of account x is the ratio of the number of logins of account x on the same IMEI of the two accounts to the number of logins of account x on all IMEIs; the degree of IMEI ownership of account y is the ratio of the number of logins of account y on the same IMEI of the two accounts to the number of logins of account y on all IMEIs. The greater the ID ownership, the more likely it is that the device corresponding to the ID belongs to the user corresponding to the account. Correspondingly, a small ID ownership usually means that the account may only occasionally log in to the device corresponding to the ID once, and there is no strong ownership relationship.

[0069] Third, determine a sub-feature vector representing the sum of the degrees of attribution of the accounts in the account pair and a sub-feature vector representing the difference in the degrees of attribution of the accounts in the account pair.

[0070] Continuing from the above example, the sum of the attribution degrees of the accounts in the account pair is and The sum of the attribution degrees of the accounts in the account pair is and difference.

[0071] Fourth, for each identification information of this type, a sub-feature vector representing the sharing degree of the identification information is determined according to the number of accounts associated with the identification information.

[0072] The sharing degree of the identification information is: the number of accounts that have logged in to the identification information. It can be understood that the higher the sharing degree of the identification information, the more likely the device corresponding to the identification information is a public device.

[0073] Step 205, input the feature vector into a pre-trained recognition model to determine whether the accounts in the account pair belong to the same person.

[0074] In this embodiment, the execution subject may input the feature vector into a pre-trained recognition model to determine whether the accounts in the account pair belong to the same person. The recognition model is used to characterize the correspondence between the feature vector corresponding to the account pair and the determination result of whether the accounts in the account pair belong to the same person.

[0075] The recognition model can be any network model with recognition function, including but not limited to convolutional neural network model, recurrent neural network model, and residual neural network model.

[0076] Continue to see Figure 3 , Figure 3 FIG. 300 is a schematic diagram of an application scenario of the method for identifying multiple accounts for one person according to this embodiment. Figure 3 In the application scenario, the account pair includes an e-commerce account x and an e-commerce account y. The server 301 first obtains the user information and the identification information respectively corresponding to the account x and the account y in the account pair from the database server 302. Then, the server determines whether there is at least one identical identification information (that is, preset association information) in the identification information involved in the account x and the account y in the account pair. Among them, the preset association information is used to characterize whether the account x and the account y in the account pair have the possibility of belonging to the same person. Then, in response to determining that the account x and the account y in the account pair have at least one identical identification information (including the second identification and the third identification), the server 301 determines whether the account x and the account y in the account pair meet the same telephone number in the receiving information, the same gender and the same age of the users corresponding to the accounts, and different ID numbers in the registration information of the accounts in the account pair (that is, the preset judgment condition). Among them, the preset judgment condition is used to determine whether the accounts in the account pair belong to the same person. Then, in response to determining that the account x and the account y in the account pair do not meet the preset judgment condition, the user information and the identification information of the account in the account pair are processed to obtain the feature vector corresponding to the account pair. Finally, the server 301 inputs the feature vector into the pre-trained recognition model to determine that the account x and account y in the account pair do not belong to the same person. The recognition model is used to characterize the correspondence between the feature vector corresponding to the account pair and the determination result of whether the accounts in the account pair belong to the same person.

[0077] The method provided by the above-mentioned embodiments of the present disclosure obtains the user information and the identification information involved corresponding to the accounts in the account pair; determines whether the accounts in the account pair satisfy the preset association information, and the preset association information is used to characterize whether the accounts in the account pair have the possibility of belonging to the same person; in response to determining that the accounts in the account pair satisfy the preset association information, determines whether the accounts in the account pair satisfy the preset judgment conditions, wherein the preset judgment conditions are used to determine whether the accounts in the account pair belong to the same person; in response to determining that the accounts in the account pair do not satisfy the preset judgment conditions, processes the user information and identification information of the accounts in the account pair to obtain a feature vector corresponding to the account pair; inputs the feature vector into a pre-trained recognition model to determine whether the accounts in the account pair belong to the same person, wherein the recognition model is used to characterize the correspondence between the feature vector corresponding to the account pair and the judgment result of whether the accounts in the account pair belong to the same person, thereby providing a method for identifying multiple accounts of one person and improving the recognition accuracy.

[0078] In some optional implementations of this embodiment, the recognition model is trained in the following manner:

[0079] First, obtain the user information and identification information corresponding to each account in the account set.

[0080] In this implementation, the account set includes a large number of accounts. The method of obtaining the user information corresponding to the account and the identification information involved can be performed with reference to the method of step 201, which will not be repeated here.

[0081] Second, the accounts in the account set that meet the preset association information are combined into account pairs to obtain multiple account pairs.

[0082] It can be understood that only accounts that meet the preset association information have the possibility of belonging to the same person. When there is a possibility that they belong to the same person, the accounts can be combined into an account pair.

[0083] Third, account pairs that meet preset judgment conditions are selected from the multiple account pairs to obtain multiple training account pairs, and labels representing whether the accounts in the training account pair belong to the same person are set for each training account pair according to the preset judgment conditions.

[0084] In this implementation, the preset judgment conditions can determine the account pairs that belong to the same person and the account pairs that do not belong to the same person. For the account pairs that belong to the same person, the above-mentioned execution subject can set a label that the accounts in the account pair belong to the same person, and use such account pairs as positive samples. For the account pairs that do not belong to the same person, the above-mentioned execution subject can set a label that the accounts in the account pair do not belong to the same person, and use such account pairs as negative samples.

[0085] Fourth, the user information and identification information of the account in each training account pair among the multiple account pairs are processed to obtain a feature vector corresponding to the training account pair.

[0086] In this implementation, the information processing method can refer to the method of step 204, which will not be repeated here.

[0087] Fifth, using machine learning methods, with the feature vectors corresponding to the training account pairs as input and the labels corresponding to the input training account pairs as expected outputs, the initial recognition model is trained to obtain a recognition model.

[0088] In this implementation, the execution subject may cyclically train the initial model through a large number of training accounts, and complete the training of the initial model in response to reaching a preset end condition, wherein the preset end condition may be, for example, the convergence of the loss function of the model and the number of training times reaching a certain number.

[0089] In some optional implementations of this embodiment, the above-mentioned execution subject can also perform information processing based on user granularity for multiple accounts belonging to the same person. Among them, user granularity is represented by users rather than accounts. As an example, information processing can be, for example, information push. When account x and account y belong to the same person, information push is performed with the users to whom account x and account y belong as the objects of information push. Specifically, information push can be performed for one account of the user. When one account of the user receives the push information, information push will no longer be performed on another account of the user. It can be understood that information processing based on user granularity can improve the user experience.

[0090] Continue to refer Figure 4 , shows a schematic process 400 of an embodiment of a method for identifying multiple accounts of one person according to the present application, comprising the following steps:

[0091] Step 401, obtaining the user information corresponding to the account in the account pair and the identification information involved.

[0092] Step 402, determining whether the accounts in the account pair satisfy the preset association information.

[0093] The preset association information is used to indicate whether the accounts in the account pair have the possibility of belonging to the same person.

[0094] Step 403, in response to determining that the account in the account pair satisfies the preset association information, determining whether the account in the account pair satisfies the first determination condition.

[0095] The first determination condition is used to determine whether the accounts in the account pair belong to the same person. When the accounts in the account pair meet the first determination condition, it is proved that the accounts in the account pair belong to the same person. It should be noted that when the accounts in the account pair do not meet the first determination condition, it cannot be proved that the accounts in the account pair do not belong to the same person.

[0096] Step 404, in response to determining that the account in the account pair does not satisfy the first determination condition, determine whether the account in the account pair satisfies the second determination condition.

[0097] The second determination condition is used to determine that the accounts in the account pair do not belong to the same person. When the accounts in the account pair meet the second determination condition, it is proved that the accounts in the account pair do not belong to the same person. It should be noted that when the accounts in the account pair do not meet the second determination condition, it cannot be proved that the accounts in the account pair belong to the same person.

[0098] Step 405, in response to determining that the account in the account pair does not satisfy the first determination condition and the second determination condition, perform the following operations:

[0099] Step 4051, based on the user information of the accounts in the account pair, determine a sub-feature vector representing the similarity of the corresponding item information between the accounts.

[0100] Step 4052: for each type of identification information, perform the following operations:

[0101] Step 40521: Determine a sub-feature vector representing the number of intersection identification information involved in the accounts in the account pair based on the identification information of the type of the accounts in the account pair.

[0102] Step 40522, for each account in the account pair, determine the sub-feature vector representing the degree of belonging of the intersection identification information relative to the account based on the number of associations between the account and the intersection identification information and the number of associations with all identification information of this type.

[0103] Step 40523, determine the sub-feature vector representing the sum of the attribution degrees of the accounts in the account pair, and the sub-feature vector representing the difference in the attribution degrees of the accounts in the account pair.

[0104] Step 40524: for each identification information of this type, determine a sub-feature vector representing the sharing degree of the identification information according to the number of accounts associated with the identification information.

[0105] Step 406, concatenate each sub-feature vector to obtain a feature vector.

[0106] Step 407, input the feature vector into a pre-trained recognition model to determine whether the accounts in the account pair belong to the same person.

[0107] Among them, the recognition model is used to characterize the correspondence between the feature vector corresponding to the account pair and the determination result of whether the accounts in the account pair belong to the same person.

[0108] It can be seen from this embodiment that Figure 2 Compared with the corresponding embodiments, the process 400 of the method for identifying multiple accounts of one person in this embodiment specifically illustrates the processing process of the feature vector and the process of determining whether the accounts of the account pair belong to the same person, further improving the recognition accuracy.

[0109] Continue to refer Figure 5 As an implementation of the methods shown in the above figures, the present disclosure provides an embodiment of a device for identifying multiple accounts of one person, and the device embodiment is Figure 2 Corresponding to the method embodiment shown, the device can be specifically applied to various electronic devices.

[0110] like Figure 5 As shown, the device for identifying multiple accounts of one person includes: including: an acquisition unit 501, configured to acquire user information corresponding to the accounts in the account pair and the identification information involved; a first determination unit 502, configured to determine whether the accounts in the account pair satisfy preset association information, and the preset association information is used to characterize whether the accounts in the account pair have the possibility of belonging to the same person; a second determination unit 503, configured to determine whether the accounts in the account pair satisfy preset determination conditions in response to determining that the accounts in the account pair satisfy the preset association information, wherein the preset determination conditions are used to determine whether the accounts in the account pair belong to the same person; a processing unit 504, configured to process the user information and identification information of the accounts in the account pair in response to determining that the accounts in the account pair do not satisfy the preset determination conditions, and obtain a feature vector corresponding to the account pair; a third determination unit 505, configured to input the feature vector into a pre-trained recognition model to determine whether the accounts in the account pair belong to the same person, wherein the recognition model is used to characterize the correspondence between the feature vector corresponding to the account pair and the determination result of whether the accounts in the account pair belong to the same person.

[0111] In some embodiments, the preset determination condition includes a first determination condition and a second determination condition, the first determination condition is used to determine that the accounts in the account pair belong to the same person, and the second determination condition is used to determine that the accounts in the account pair do not belong to the same person.

[0112] In some embodiments, the processing unit 504 is further configured to: process corresponding item information between accounts in the account pair to obtain a sub-feature vector corresponding to each corresponding item information; and concatenate each sub-feature vector to obtain a feature vector.

[0113] In some embodiments, the user information includes user portrait information, consumption habit information and receipt information; the processing unit 504 is further configured to: determine a sub-feature vector representing the similarity of corresponding item information between the accounts based on the user information of the accounts in the account pair.

[0114] In some embodiments, the processing unit 504 is further configured to: perform the following operations for each type of identification information: determine a sub-feature vector representing the number of intersection identification information involved in all accounts in the account pair based on the identification information of the type of the account in the account pair; for each account in the account pair, determine a sub-feature vector representing the degree of belonging of the intersection identification information relative to the account based on the number of associations between the account and the intersection identification information and the number of associations with all identification information of the type; determine a sub-feature vector representing the sum of the degrees of belonging of the accounts in the account pair and a sub-feature vector representing the difference in the degrees of belonging of the accounts in the account pair; for each identification information of the type, determine a sub-feature vector representing the degree of sharing of the identification information based on the number of accounts associated with the identification information.

[0115] In some embodiments, the above-mentioned device also includes: a training unit (not shown in the figure), which is configured to train and obtain a recognition model in the following manner: obtain the user information and the identification information involved corresponding to each account in the account set; combine the accounts in the account set that meet the preset association information into account pairs to obtain multiple account pairs; screen out account pairs that meet the preset judgment conditions from the multiple account pairs to obtain multiple training account pairs, and set a label for each training account pair according to the preset judgment conditions to represent whether the accounts in the training account pair belong to the same person; process the user information and identification information of the account in each training account pair in the multiple account pairs to obtain a feature vector corresponding to the training account pair; use a machine learning method, with the feature vector corresponding to the training account pair as input, and the label corresponding to the input training account pair as the expected output, to train the initial recognition model to obtain a recognition model.

[0116] In this embodiment, an acquisition unit in a device for identifying multiple accounts of one person acquires user information and identification information corresponding to the accounts in an account pair; a first determination unit determines whether the accounts in the account pair satisfy preset association information, and the preset association information is used to characterize whether the accounts in the account pair have the possibility of belonging to the same person; a second determination unit determines whether the accounts in the account pair satisfy preset judgment conditions in response to determining that the accounts in the account pair satisfy the preset association information, wherein the preset judgment conditions are used to determine whether the accounts in the account pair belong to the same person; a processing unit processes the user information and identification information of the accounts in the account pair in response to determining that the accounts in the account pair do not satisfy the preset judgment conditions, and obtains a feature vector corresponding to the account pair; a third determination unit inputs the feature vector into a pre-trained recognition model to determine whether the accounts in the account pair belong to the same person, wherein the recognition model is used to characterize the correspondence between the feature vector corresponding to the account pair and the judgment result of whether the accounts in the account pair belong to the same person, thereby providing a device for identifying multiple accounts of one person and improving the recognition accuracy.

[0117] Reference below Figure 6 , which shows a device suitable for implementing the embodiments of the present application (eg Figure 1 A schematic diagram of the structure of a computer system 600 of the devices 101, 102, 103, 105 shown. Figure 6 The device shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0118] like Figure 6 As shown, the computer system 600 includes a processor (e.g., CPU, central processing unit) 601, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 602 or a program loaded from a storage part 608 into a random access memory (RAM) 603. Various programs and data required for the operation of the system 600 are also stored in the RAM 603. The processor 601, the ROM 602, and the RAM 603 are connected to each other via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0119] The following components are connected to the I / O interface 605: an input section 606 including a keyboard, a mouse, etc.; an output section 607 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 608 including a hard disk, etc.; and a communication section 609 including a network interface card such as a LAN card, a modem, etc. The communication section 609 performs communication processing via a network such as the Internet. A drive 610 is also connected to the I / O interface 605 as needed. A removable medium 611, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 610 as needed, so that a computer program read therefrom is installed into the storage section 608 as needed.

[0120] In particular, according to an embodiment of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program contains a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through the communication part 609, and / or installed from a removable medium 611. When the computer program is executed by the processor 601, the above-mentioned functions defined in the method of the present application are executed.

[0121] It should be noted that the computer-readable medium of the present application may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present application, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0122] Computer program code for performing the operations of the present application may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" or similar programming languages. The program code may be executed entirely on the client computer, partially on the client computer, as a separate software package, partially on the client computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the client computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., through the Internet using an Internet service provider).

[0123] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the device, method and computer program product according to various embodiments of the present application. In this regard, each box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flow chart, and the combination of the boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0124] The units involved in the embodiments described in the present application may be implemented by software or by hardware. The units described may also be set in a processor, for example, may be described as: a processor comprising an acquisition unit, a first determination unit, a second determination unit, a processing unit, and a third determination unit. Among them, the names of these units do not constitute a limitation on the unit itself under certain circumstances. For example, the third determination unit may also be described as "a unit that inputs a feature vector into a pre-trained recognition model to determine whether the accounts in the account pair belong to the same person."

[0125] As another aspect, the present application also provides a computer-readable medium, which may be included in the device described in the above embodiment; or may exist independently and not assembled into the device. The above computer-readable medium carries one or more programs, and when the above one or more programs are executed by the device, the computer device: obtains the user information corresponding to the account in the account pair and the identification information involved; determines whether the account in the account pair satisfies the preset association information, and the preset association information is used to characterize whether the account in the account pair has the possibility of belonging to the same person; in response to determining that the account in the account pair satisfies the preset association information, determines whether the account in the account pair satisfies the preset judgment condition, wherein the preset judgment condition is used to determine whether the account in the account pair belongs to the same person; in response to determining that the account in the account pair does not satisfy the preset judgment condition, processes the user information and identification information of the account in the account pair to obtain the feature vector corresponding to the account pair; inputs the feature vector into the pre-trained recognition model to determine whether the account in the account pair belongs to the same person, wherein the recognition model is used to characterize the corresponding relationship between the feature vector corresponding to the account pair and the judgment result of whether the account in the account pair belongs to the same person.

[0126] The above description is only a preferred embodiment of the present application and an explanation of the technical principles used. Those skilled in the art should understand that the scope of the invention involved in the present application is not limited to the technical solution formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above invention concept. For example, the above features are replaced with the technical features with similar functions disclosed in this application (but not limited to) by each other to form a technical solution.

Claims

1. A method for identifying multiple accounts of one person, comprising: Obtain the user information corresponding to the account in the account pair and the identification information involved; Determining whether the accounts in the account pair satisfy preset association information, where the preset association information is used to indicate whether the accounts in the account pair have a possibility of belonging to the same person; In response to determining that the accounts in the account pair satisfy the preset association information, determining whether the accounts in the account pair satisfy a preset determination condition, wherein the preset determination condition is used to determine whether the accounts in the account pair belong to the same person; In response to determining that the account in the account pair does not meet the preset judgment condition, the user information and identification information of the account in the account pair are processed to obtain a feature vector corresponding to the account pair, including: for each type of identification information, the following operations are performed: according to the identification information of the type of the account in the account pair, a sub-feature vector representing the number of intersection identification information involved in the accounts in the account pair is determined; for each account in the account pair, according to the number of associations between the account and the intersection identification information and the number of associations with all identification information of the type, a sub-feature vector representing the degree of belonging of the intersection identification information relative to the account is determined; a sub-feature vector representing the sum of the degrees of belonging of the accounts in the account pair and a sub-feature vector representing the difference in the degrees of belonging of the accounts in the account pair are determined; for each identification information of the type, a sub-feature vector representing the degree of sharing of the identification information is determined according to the number of accounts associated with the identification information; each sub-feature vector is concatenated to obtain the feature vector; The feature vector is input into a pre-trained recognition model to determine whether the accounts in the account pair belong to the same person, wherein the recognition model is used to characterize the correspondence between the feature vector corresponding to the account pair and the determination result of whether the accounts in the account pair belong to the same person.

2. The method according to claim 1, wherein: The preset determination condition includes a first determination condition and a second determination condition, wherein the first determination condition is used to determine that the accounts in the account pair belong to the same person, and the second determination condition is used to determine that the accounts in the account pair do not belong to the same person.

3. The method according to claim 1, wherein: The processing of the user information and identification information of the account in the account pair to obtain a feature vector corresponding to the account pair includes: Processing corresponding item information between the accounts in the account pair to obtain a sub-feature vector corresponding to each corresponding item information; Each sub-feature vector is concatenated to obtain the feature vector.

4. The method according to claim 3, wherein: The user information includes user portrait information, consumption habit information and receipt information; The processing of the corresponding item information between the accounts in the account pair to obtain a sub-feature vector corresponding to each corresponding item information includes: Based on the user information of the accounts in the account pair, a sub-feature vector representing the similarity of the corresponding item information between the accounts is determined.

5. The method according to any one of claims 1 to 4, wherein: The recognition model is trained in the following way: Obtain the user information and identification information corresponding to each account in the account set; Combining accounts in the account set that meet the preset association information into account pairs to obtain multiple account pairs; Filtering account pairs that meet the preset determination condition from the multiple account pairs to obtain multiple training account pairs, and setting a label for each training account pair according to the preset determination condition to indicate whether the accounts in the training account pair belong to the same person; Processing user information and identification information of an account in each training account pair among the multiple account pairs to obtain a feature vector corresponding to the training account pair; The recognition model is obtained by using a machine learning method, taking the feature vector corresponding to the training account pair as input and the label corresponding to the input training account pair as the expected output, training the initial recognition model.

6. A device for identifying multiple accounts of one person, comprising: an acquisition unit configured to acquire user information corresponding to an account in the account pair and the identification information involved; A first determining unit is configured to determine whether the accounts in the account pair satisfy preset association information, where the preset association information is used to indicate whether the accounts in the account pair have a possibility of belonging to the same person; a second determining unit configured to determine, in response to determining that the accounts in the account pair satisfy the preset association information, whether the accounts in the account pair satisfy a preset determination condition, wherein the preset determination condition is used to determine whether the accounts in the account pair belong to the same person; The processing unit is configured to process the user information and identification information of the account in the account pair in response to determining that the account in the account pair does not meet the preset judgment condition, and obtain the feature vector corresponding to the account pair, including: for each type of identification information, perform the following operations: according to the identification information of the type of the account in the account pair, determine a sub-feature vector representing the number of intersection identification information involved in the accounts in the account pair; for each account in the account pair, according to the number of associations between the account and the intersection identification information and the number of associations with all identification information of the type, determine a sub-feature vector representing the degree of belonging of the intersection identification information relative to the account; determine a sub-feature vector representing the sum of the degrees of belonging of the accounts in the account pair, and a sub-feature vector representing the difference in the degrees of belonging of the accounts in the account pair; for each identification information in the type, determine a sub-feature vector representing the degree of sharing of the identification information according to the number of accounts associated with the identification information; splice each sub-feature vector to obtain the feature vector; The third determination unit is configured to input the feature vector into a pre-trained recognition model to determine whether the accounts in the account pair belong to the same person, wherein the recognition model is used to characterize the correspondence between the feature vector corresponding to the account pair and the determination result of whether the accounts in the account pair belong to the same person.

7. The device according to claim 6, wherein: The preset determination condition includes a first determination condition and a second determination condition, wherein the first determination condition is used to determine that the accounts in the account pair belong to the same person, and the second determination condition is used to determine that the accounts in the account pair do not belong to the same person.

8. The device according to claim 6, wherein: The processing unit is further configured to: Corresponding item information between the accounts in the account pair is processed to obtain a sub-feature vector corresponding to each corresponding item information; and each sub-feature vector is concatenated to obtain the feature vector.

9. The device according to claim 8, wherein: The user information includes user portrait information, consumption habit information and receipt information; The processing unit is further configured to: Based on the user information of the accounts in the account pair, a sub-feature vector representing the similarity of the corresponding item information between the accounts is determined.

10. The device according to any one of claims 6 to 9, wherein: Also includes: The training unit is configured to train the recognition model in the following manner: obtaining user information and identification information corresponding to each account in the account set; combining the accounts in the account set that meet the preset association information into account pairs to obtain multiple account pairs; Filtering account pairs that meet the preset determination condition from the multiple account pairs to obtain multiple training account pairs, and setting a label for each training account pair according to the preset determination condition to indicate whether the accounts in the training account pair belong to the same person; Processing the user information and identification information of each account in each training account pair among the plurality of account pairs to obtain a feature vector corresponding to the training account pair; using The machine learning method uses the feature vector corresponding to the training account pair as input, uses the label corresponding to the input training account pair as the expected output, trains the initial recognition model, and obtains the recognition model.

11. A computer readable medium having a computer program stored thereon, wherein: When the program is executed by a processor, the method according to any one of claims 1 to 5 is implemented.

12. An electronic device comprising: one or more processors; a storage device having one or more programs stored thereon, When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 5.

Images