User category identification method, device, electronic device and readable storage medium

Abstract

The embodiment of the present application provides a user category identification method, device, electronic device and readable storage medium, the method comprising: obtaining target device information of a terminal device used by a user, and obtaining target network environment information of a network environment where the terminal device is located; and identifying the target user category to which the user belongs according to the target device information and the target network environment information. The method can identify the target user category to which the user belongs in real time and accurately.

Description

Technical Field

[0001] The embodiments of the present disclosure relate to the field of Internet technology, and more specifically, to a user category identification method, device, electronic device, and computer-readable storage medium. Background Art

[0002] With the development of Internet technology, transactions through online platforms have become a common behavior among Internet users. However, while transactions based on online platforms bring convenience to users, they may also bring security risks to normal users due to the existence of black and gray industry users. In order to maintain the ecological health of online platforms and ensure a good user experience, how to identify black and gray industry users in real time and accurately has become an urgent problem to be solved.

[0003] At present, the user category to which the user belongs is generally identified based on the user's historical behavior characteristics to determine whether the user is a black or gray market user. This method has a strong dependence on the user's historical behavior characteristics, which means that when the user does not have the user's historical behavior characteristics or does not have the user's historical behavior characteristics within the evaluation period, the network platform often cannot determine the user category in real time and accurately. Summary of the invention

[0004] One purpose of the present disclosure is to provide a new technical solution for identifying user categories, so as to solve the problem that the prior art may not be able to identify the user category of a user in real time and accurately.

[0005] In a first aspect of the present disclosure, a method for identifying a user category is provided, the method comprising:

[0006] Acquire target device information of a terminal device used by a user, and acquire target network environment information of a network environment in which the terminal device is located;

[0007] According to the target device information and the target network environment information, a target user category to which the user belongs is identified.

[0008] According to a second aspect of the present disclosure, a user category identification device is provided, the device comprising:

[0009] An acquisition module, used to acquire target device information of a terminal device used by a user, and to acquire target network environment information of a network environment where the terminal device is located;

[0010] The identification module is used to identify the target user category to which the user belongs according to the target device information and the target network environment information.

[0011] According to a third aspect of the present disclosure, there is also provided an electronic device, including:

[0012] A memory for storing executable instructions;

[0013] The processor is used to operate the electronic device to execute the method described in the first aspect of the present disclosure under the control of the instruction.

[0014] According to a fourth aspect of the present disclosure, there is further provided a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program that can be read and executed by a computer, and the computer program is used to execute the method according to the first aspect of the present disclosure when read and executed by the computer.

[0015] One beneficial effect of the embodiments of the present disclosure is that according to the embodiments of the present disclosure, when the network platform identifies the user category to determine whether the user is a black and gray market user, it is not necessary to rely on the user's historical behavioral characteristics. Instead, with the user's authorization, it can obtain the target device information of the terminal device used by the user to log in or access the network platform, and obtain the target network environment information of the network environment in which the terminal device is located. Based on the target device information and the target network environment information, the target user category to which the user belongs can be identified in real time and accurately.

[0016] Other features and advantages of the present specification will become apparent from the following detailed description of exemplary embodiments of the present specification with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The accompanying drawings, which are incorporated in and constitute a part of the specification, illustrate embodiments of the specification and, together with the description, serve to explain the principles of the specification.

[0018] Figure 1 It is a flowchart of the user category identification method provided by an embodiment of the present disclosure.

[0019] Figure 2 It is a schematic diagram of a target graph network provided by an embodiment of the present disclosure.

[0020] Figure 3 is a flowchart illustrating a user category identification process.

[0021] Figure 4 It is a principle block diagram of a user category identification device provided by an embodiment of the present disclosure.

[0022] Figure 5 It is a schematic diagram of the hardware structure of the electronic device provided by the embodiment of the present disclosure. DETAILED DESCRIPTION

[0023] Various exemplary embodiments of the present disclosure will now be described in detail with reference to the accompanying drawings. It should be noted that the relative arrangement of components and steps, numerical expressions and numerical values ​​set forth in these embodiments do not limit the scope of the present invention unless otherwise specifically stated.

[0024] The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the invention, its application, or uses.

[0025] Technologies, methods, and equipment known to ordinary technicians in the relevant art may not be discussed in detail, but where appropriate, the technologies, methods, and equipment should be considered as part of the specification.

[0026] In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not limiting. Therefore, other examples of the exemplary embodiments may have different values.

[0027] It should be noted that like reference numerals and letters refer to similar items in the following figures, and therefore, once an item is defined in one figure, it need not be further discussed in subsequent figures.

[0028] <Method Example>

[0029] In the prior art, when identifying whether a user is a user of the black and gray production category, the user is generally scored based on the frequency of the user's historical behavior, and the user category is determined based on the scoring results. This type of method specifically counts the frequency of user behavior in various business scenarios within a certain period of time, such as the number of recharges, the number of logins, the number of speeches, etc.; then the frequency is used to calculate relevant indicators, such as user activity, or the user score is directly calculated through a model algorithm to determine the user category. In addition, there are also methods that score users based on the user's historical behavior sequence and determine the user category based on the scoring results. This type of method specifically collects the time series sequence of the user's behavior generated within a certain period of time, then analyzes and processes the behavior sequence data, and calculates the user score through an algorithm model to determine the user category.

[0030] It can be seen that the existing methods are highly dependent on the user's historical behavior characteristics when identifying user categories. Such methods cannot accurately identify users without historical behavior characteristics, such as new users, or users without historical behavior characteristics within the statistical period, such as returning users. In addition, the method of identifying user categories based on user behavior data collected and counted over a period of time still has a certain lag because it takes time to collect user behavior data, and there is a problem of not being able to identify user categories in real time.

[0031] To solve the above problems, the present invention provides a method for identifying user categories. Figure 1 , which is a flow chart of the user category identification method provided in the embodiment of the present disclosure. The method can be implemented by an electronic device, and the electronic device can be a server running a service end of a network platform, and the server can be a blade server, a rack server, etc.; or, the electronic device can also be a terminal device running a client of a network platform, and the terminal device can be a mobile phone, a platform computer, etc. used by the user. When the user logs in or accesses the network platform based on the terminal device, the client running in the terminal device executes the method to determine the target user category to which the user belongs, and sends the target user category to the server, so that the server can correctly respond to the user's business behavior in business key nodes, such as payment, resource exchange and other business scenarios according to the target user category. In the embodiment of the present disclosure, for the convenience of description, the electronic device is taken as an example as a server.

[0032] like Figure 1 The method of this embodiment may include the following steps S1100-S1200, which are described in detail below.

[0033] Step S1100, obtaining target device information of a terminal device used by a user, and obtaining target network environment information of a network environment in which the terminal device is located.

[0034] Target device information refers to information associated with the terminal device used by the user. The target device information may include device attribute information used to describe device attributes, such as device identification, device model identification, client version identification, etc., wherein the client version identification is used to indicate the version of the client of the network platform installed in the terminal device.

[0035] In addition, the target device information can also be cheating feature information obtained by detecting the internal software environment of the terminal device, and the cheating feature information can be information indicating whether the user has a motive to tamper with the internal software environment of the terminal device. The cheating feature information can be, for example, at least one of the first feature information indicating whether the terminal device is rooted, the second feature information indicating whether the terminal device is iOS jailbreaking, the third feature information indicating whether the terminal device is hooked, and the fourth feature information indicating whether the terminal device is injected, wherein whether the terminal device is rooted refers to whether the device running the Android operating system obtains the highest access rights; whether the terminal device is jailbroken refers to whether the device running the iOS operating system obtains the highest access rights; whether the terminal device is hooked refers to whether the hook technology is used to hijack the original function in the system; whether the terminal device is injected refers to whether the device uses the injection technology to abnormally access and tamper with the information in the target process space.

[0036] It should be noted that, in the embodiments of the present disclosure, the above-mentioned cheating feature information is used to describe the internal software environment in the mobile terminal device. In specific implementation, the cheating feature information can also be used to describe the internal software environment in the non-mobile terminal device. For example, it can be feature information used to describe whether a computer running a Windows operating system has administrator privileges. Of course, the cheating feature information can also be other feature information other than the above-mentioned first feature information, second feature information, third feature information and fourth feature information, which is not specifically limited here.

[0037] The target network environment information is information indicating the network environment in which the terminal device is located. The target network environment information may include information such as the Internet Protocol address (IP, Internet Protocol Address), the network type, etc., wherein the network type may be information used to describe whether the network environment is a local area network or a wireless network.

[0038] In specific implementation, the target device information and target network environment information can be obtained in real time when the user registers or logs in to the network platform using the client application corresponding to the network platform in the terminal device, with the user's authorization, so as to ensure the real-time and accuracy of the two types of information, thereby improving the real-time and accuracy of the final recognition result.

[0039] Step S1200: identifying the target user category to which the user belongs according to the target device information and the target network environment information.

[0040] After obtaining the target device information of the terminal device used by the user to access the network platform and the target network environment information of the network environment in which the terminal device is located through the above step S1100, the user category can be identified and processed based on the two types of information. The following describes how to identify the target user category to which the user belongs.

[0041] In one embodiment, the identifying the target user category to which the user belongs based on the target device information and the target network environment information includes: constructing a target graph network representing the association relationship between the user and other users based on the target device information and the target network environment information; detecting cheating feature information of the user based on the target device information; and identifying the target user category to which the user belongs based on the target graph network and the cheating feature information.

[0042] The target graph network is a network topology structure composed of multiple graph nodes and connection edges for connecting different graph nodes. In the graph network, two graph nodes can be connected by using connection edges to represent the association relationship between the two graph nodes, such as dependency relationship, inclusion relationship, etc.

[0043] The cheating feature information of the user may be a user vector of whether the user has the corresponding cheating feature information. For example, if the value "0" indicates that the user has the corresponding cheating feature information, and the value "1" indicates that the user does not have the corresponding cheating feature information, then when it is detected that the terminal device used by user 1 is rooted, not jailbroken, not hooked, and has been injected, a user vector (0,1,1,0) may be constructed to identify the user vector of user 1, wherein the value of each element in the user vector is determined according to whether the user has the corresponding cheating feature information.

[0044] Generally, according to the degree of connection between different graph nodes in a graph network, the graph network can also be regarded as consisting of different clusters. The graph nodes within a cluster often have closer connections, while the connections between different clusters are relatively sparse. This gives rise to the concept of community, that is, a community in a graph network refers to a set composed of multiple tightly connected graph nodes. Multiple graph nodes in the same community have more internal connections and relatively fewer external connections.

[0045] Specifically in the embodiments of the present disclosure, based on the fact that black and gray market behaviors generally correspond to group behaviors, therefore, multiple graph nodes corresponding to the user to be identified can be generated based on the user's target device information and target network environment information, and a target graph network can be constructed based on the multiple graph nodes and multiple graph nodes corresponding to other users to determine the target community to which the user belongs, and based on the data in the target community, the target user classification of the user to be identified can be identified.

[0046] Please see Figure 1, which is a schematic table showing the users who access the network platform and the device information corresponding to each user, wherein, in Table 1, Uid is the user identifier, which is used to uniquely represent a user; IP is used to represent the IP address of the terminal device used by the corresponding user; DeviceId is the terminal device identifier, which is used to uniquely represent a terminal device; DeviceModel is the device model identifier used to represent the device model of the corresponding terminal device; ClientVer is the client version identifier, which is used to represent the version of the client of the network platform installed in the terminal device.

[0047] Uid IP DeviceId DeviceModel ClientVer 1 127.0.0.1 a4cc3f Model 1 1.0.1 2 127.0.0.1 a4cc3f Model 1 1.0.1 3 127.0.0.1 f4ecdb Model 2 1.0.2 4 127.0.0.2 f4ecdb Model 2 1.0.2 5 127.0.0.2 a4cc3f Model 1 1.0.1 6 127.1.1.1 eccdf4 Model 1 1.0.3 7 127.1.1.1 eccdf4 Model 1 1.0.3

[0048] Table 1

[0049] As shown in Table 1, the specific value of each Uid is used as a user node in the graph network, the specific value of each IP address is used as an IP node in the graph network, the specific value of each device identifier is used as a device node in the graph network, the specific value of each device model identifier is used as a device model node in the graph network, and the client identifier is used as a client version node in the graph network. According to the IP address, device identifier, device model and client identifier corresponding to each user shown in Table 1, the connection edges between the corresponding user node and the corresponding IP node, the corresponding device node, the corresponding device model node and the corresponding client version identifier node in the graph network are established, and the following can be constructed: Figure 2 The target graph network shown.

[0050] After constructing the target group network corresponding to the user and detecting the cheating feature information corresponding to the user, the target user category to which the user belongs can be identified based on the target graph network and the cheating feature information.

[0051] Please see Figure 3 , which is a flowchart illustrating the user category identification process. Figure 3 As shown, in one embodiment, identifying the target user category to which the user belongs based on the target graph network and the cheating feature information may include the following steps S1210-S1240.

[0052] Step S1210: Use a preset community algorithm to perform community analysis on the target graph network to obtain community data corresponding to the user.

[0053] According to the above steps of constructing the target graph network, it can be known that the target graph network corresponding to the user constructed by the embodiments of the present disclosure includes multiple graph nodes, and the connection relationship between the multiple graph nodes is determined according to the device information and network environment information of the corresponding user; the use of a preset community algorithm to perform community analysis on the target graph network to obtain community data corresponding to the user includes: according to the connection relationship between multiple graph nodes in the target graph network, using a preset community algorithm to perform community analysis, determine the target community to which the user belongs, and obtain data in the target community as the community data.

[0054] The preset community algorithm may be any one of the algorithms such as the GN algorithm, the label propagation algorithm (LabelPropagation), the random walk algorithm, the FastUnfolding algorithm, and the BGLL algorithm, and is not specifically limited here.

[0055] according to Figure 2 From the target graph network shown, it can be seen that for the user to be identified with Uid "1", that is, user 1, the users with Uids "1", "2", "3", "4" and "5" are obviously a community, so this community can be regarded as the target community of user 1, and the data in the target community can be used as the community data of user 1.

[0056] according to Figure 2 As shown, the community data contains a total of 13 graph nodes, including 5 user nodes and 8 device environment information nodes corresponding to user devices and networks. The community data can be specifically represented by the tuple (5, (5-1) / 13), wherein the first element in the tuple represents the total number of user nodes in the corresponding community, and the second element represents the proportion of user nodes in the community summary points, that is, the proportion of other user nodes in the community summary points after removing the user node with a minimum value of 1.

[0057] Step S1220: determining a user vector of the user according to the cheating feature information.

[0058] In an embodiment of the present disclosure, a user vector of a user may include multiple feature elements, and the values ​​of corresponding feature elements are used to indicate whether the terminal device used by the user has corresponding cheating feature information.

[0059] For ease of understanding, the cheating characteristic information is taken as the first characteristic information, the second characteristic information, the third characteristic information and the fourth characteristic information in the above description, that is, root, iOS jailbreak, hook and injection for example.

[0060] Please refer to Table 2, which is a table showing the cheating feature information corresponding to the terminal devices used by the users with UIDs "1", "2", "3", "4" and "5" in the target community.

[0061]

[0062]

[0063] Table 2

[0064] According to Table 2, we can construct that the user vector of user 1 with Uid "1" is (0,1,1,0), the user vector of user 2 with Uid "2" is (0,1,1,0), the user vector of user 3 with Uid "3" is (1,0,0,1), the user vector of user 4 with Uid "4" is (1,1,1,1), and the user vector of user 5 with Uid "5" is (0,1,1,0).

[0065] Step S1230: Calculate the user rating of the user based on the community data and the user vector.

[0066] Specifically, according to the above description, in an embodiment of the present disclosure, the user vector of the user to be identified may include multiple feature elements, and the value of each feature element is determined according to whether the user has corresponding cheating feature information; in a specific implementation, the user score of the user is calculated based on the community data and the user vector, which may include the following steps S1231-S1233.

[0067] Step S1231, obtaining a target weight vector corresponding to the user vector, wherein the target weight vector includes weight elements corresponding to each feature element in the user vector.

[0068] Specifically, after constructing the user vector corresponding to the user, since the degree of harm of each cheating feature information is often different, for example, the harm of hook is obviously higher than that of root. Therefore, when constructing the user vector and calculating the user score based on the user vector to determine the target user classification, it is also necessary to determine the weight of each feature element in the user vector, so as to obtain a more accurate user score in a targeted manner through weighted averaging, thereby improving the accuracy of the final recognition result.

[0069] In an embodiment of the present disclosure, in order to make the user rating and the target user classification determined based on the user rating have strong interpretability, the target weight vector can be determined by constructing a judgment matrix. Specifically, the target weight vector can be determined by the following steps: constructing a first judgment matrix according to the relative importance of all cheating feature information corresponding to the user; normalizing the first element in each column of the first judgment matrix to obtain a second judgment matrix; summing the second elements in each row of the second judgment matrix to obtain a third judgment matrix; obtaining an initial weight vector by calculating the ratio between each third element and the sum of all third elements in the third judgment matrix; and obtaining the target weight vector based on the initial weight vector.

[0070] The judgment matrix refers to comparing any two evaluation indicators to obtain an importance result, and then presenting this result in the form of a matrix, where the degree of importance difference can be represented by Table 3.

[0071] Determine the importance meaning 1 Two comparative evaluation indicators are of equal importance 3 Two comparative evaluation indicators, the former is slightly more important than the latter 5 Two comparative evaluation indicators, the former is more important than the latter 7 Two comparative evaluation indicators, the former is obviously more important than the latter 9 Two comparative evaluation indicators, the former is more important than the latter 2，4，6，8 Indicates the middle value of the judgment results of two evaluation indicators

[0072] Table 3

[0073] As shown in Table 3, when measuring two evaluation indicators in the judgment matrix, the larger the value, the more important the former is than the latter, and the importance of the two evaluation indicators is the reciprocal of each other. For example, for evaluation indicators A and B, when the importance of A compared to B is 5, the importance of B compared to A is 1 / 5.

[0074] The following takes the first characteristic information, the second characteristic information, the third characteristic information and the fourth characteristic information corresponding to the terminal device as an example, i.e., root, iOS jailbreak, hook and injection. Figure 3 Taking the judgment importance as an example, it explains how to construct a judgment matrix to determine the target weight vector.

[0075] Specifically, the first judgment matrix Matrix1 as shown in the following Table 4 can be constructed according to the relative importance between root, iOS jailbreak, hook and injection.

[0076] root iOS Jailbreak hook injection root 1 1 1 / 5 1 / 3 iOS Jailbreak 1 1 1 / 5 1 / 3 hook 5 5 1 5 / 3 injection 3 3 3 / 5 1

[0077] Table 4

[0078] After the first judgment matrix is ​​constructed, the first elements in the first judgment matrix are normalized by column. The normalization process can be: calculating the sum of all the first elements included in each column, and setting the value of each element in the column to the ratio of the corresponding first element to the sum to complete the normalization process. For example, if the elements in a column are A1, A2, A3, and A4, then when the sum of A1, A2, A3, and A4 is SUM1, the normalized elements of the column are A1 / SUM1, A2 / SUM1, A3 / SUM1, and A4 / SUM1.

[0079] According to the above description, after the first judgment matrix shown in Table 4 is normalized by column, the second judgment matrix Matrix2 shown in Table 5 below can be obtained.

[0080]

[0081]

[0082] Table 5

[0083] After the second judgment matrix shown in Table 5 is constructed, by summing the second elements of each row in the second judgment matrix, a third judgment matrix Matrix3 shown in Table 6 below can be obtained.

[0084] root 2 / 5 iOS Jailbreak 2 / 5 hook 2 injection 6 / 5

[0085] Table 6

[0086] Afterwards, by calculating the ratio between each third element and the sum of all third elements in the third judgment matrix, an initial weight vector as shown in the following Table 7 can be obtained.

[0087] root 1 / 10 iOS Jailbreak 1 / 10 hook 1 / 2 injection 3 / 10

[0088] Table 7

[0089] That is, the initial weight vector corresponding to each user vector can be obtained as (1 / 10, 1 / 10, 1 / 2, 3 / 10).

[0090] It should be noted that after the initial weight vector is constructed, the target weight vector is obtained based on the initial weight vector, including: if the initial weight vector is subjected to a consistency check and the check passes, determining that the initial weight vector is the target weight vector; if the initial weight vector is subjected to a consistency check and the check fails, reconstructing the first judgment matrix to determine the target weight vector.

[0091] Specifically, after constructing the initial weight vector, in order to determine whether the initial weight vector is accurate and valid, it is also necessary to perform a consistency check on the initial weight vector. The consistency check step can be specifically: calculating the maximum eigenvalue root of the first judgment matrix initially constructed; calculating the consistency index CI; determining the random consistency index based on the number of evaluation indicators; and calculating the consistency ratio.

[0092] For example, when the first judgment matrix is ​​the above Matrix1 and the initial weight vector is the above W, the maximum eigenvalue of the first judgment matrix can be calculated by the following steps:

[0093]

[0094]

[0095] After calculating the above maximum characteristic root, the consistency index CI can be calculated by the following steps:

[0096]

[0097] Wherein, n is the number of evaluation indicators. In this example, since the evaluation indicators are root, iOS jailbreak, hook and injection, n is 4.

[0098] In addition, according to the value of n, the random consistency index RI can be determined to be 0.9 by looking up the table; then, through the consistency ratio CR=CI / RI, it can be obtained that CR=0 / 0.9=0. Since 0<0.9, it can be determined that the initial weight vector meets the consistency check and can be determined to be the target weight vector.

[0099] It should be noted that in the embodiments of the present disclosure, the target weight vector corresponding to the user vector is determined by constructing a judgment matrix. In specific implementation, the target weight vector can of course also be determined based on a neural network model. However, the steps of determining the weight of each feature element in the user vector based on the neural network model are invisible to the user, which causes the final determined target weight vector to be unexplainable, and thus the final recognition result may be unexplainable. By determining the target weight matrix based on the above method of constructing a judgment matrix, on the one hand, the recognition result can be more accurate, and on the other hand, the recognition result can be highly interpretable.

[0100] Step S1232, based on the community data, obtain the total number of user nodes in the target community to which the user belongs, and obtain the percentage of the number of user nodes corresponding to the user, wherein the percentage of the number of user nodes is a ratio between a first number and a second number of all nodes in the target community, and the first number is the number of all nodes that have a direct connection relationship with the user's user node; and, step S1233, divide the product of the user vector, the target weight vector and the percentage of the number of user nodes by the total number of user nodes to obtain a user score.

[0101] In an embodiment of the present disclosure, the user rating can be calculated by the formula: Calculated, where a represents the user vector, w represents the target weight vector, c is the total number of user nodes in the target community corresponding to the user, and b is the ratio between the first number of all user nodes in the target community and the second number of all nodes.

[0102] Specifically, when user 1, i.e., the user vector of user with Uid "1" is (0, 1, 1, 0), the target weight vector is (1 / 10, 1 / 10, 1 / 2, 3 / 10), the number of user nodes in the target community is 5, and the ratio between the first number of all user nodes and the second number of all nodes is 5 / 13, the user score of user 1 is

[0103] Step S1240: Determine the target user category based on the user score.

[0104] After calculating the user score corresponding to the user to be identified according to the above steps, the target user category of the user can be determined according to the user score. For example, when the user score is greater than a preset threshold, the user can be determined to be a black and gray industry user.

[0105] After determining the target user category of the user to be identified according to the above steps, the user behavior can be correctly responded to in the key business nodes based on the target user category to improve the security of the network platform.

[0106] It should be noted that in the embodiments of the present disclosure, the user categories of black and gray production categories and non-black and gray production categories are taken as examples, that is, the final identification and determination of whether the user is a black and gray production user is used as an example for explanation. In specific implementation, the method can also be applied to scenarios for identifying other user categories. For example, users are classified according to their consumption enthusiasm, and by identifying whether the user is a user in the active consumption category, different interactive events, virtual resources, etc. are recommended to the user in a targeted manner to improve user stickiness. In addition, it should be noted that when calculating the user score above to determine the target user category, the user's historical behavior characteristics can also be introduced if the user has the corresponding historical behavior characteristics to further improve the accuracy of the recognition result. The detailed processing steps will not be repeated here.

[0107] To sum up, the user category identification method provided by the embodiments of the present disclosure is that when the network platform identifies the user category to determine whether the user is a black and gray market user, it does not need to rely on the user's historical behavioral characteristics. Instead, it can obtain the target device information of the terminal device used by the user to log in or access the network platform, and obtain the target network environment information of the network environment in which the terminal device is located, with the user's authorization. Based on the target device information and the target network environment information, the target user category to which the user belongs can be identified in real time and accurately.

[0108] <Device Example>

[0109] Corresponding to the above method embodiment, in this embodiment, a user category identification device is also provided, such as Figure 4 As described above, the device 400 may include an acquisition module 410 and an identification module 420 .

[0110] The acquisition module 410 is used to obtain the target device information of the terminal device used by the user, and to obtain the target network environment information of the network environment in which the terminal device is located; the identification module 420 is used to identify the target user category to which the user belongs based on the target device information and the target network environment information.

[0111] <Equipment Embodiment>

[0112] In this embodiment, an electronic device is also provided. Figure 5 As described above, the electronic device 500 may include a processor 520 and a memory 510, wherein the memory 510 is used to store executable instructions; the processor 520 is used to operate the electronic device according to the control of the instructions to execute the method according to any embodiment of the present disclosure.

[0113] <Computer Readable Storage Medium Embodiment>

[0114] This embodiment provides a computer-readable storage medium, in which executable commands are stored. When the executable commands are executed by a processor, the method described in any method embodiment of this specification is executed.

[0115] The flowcharts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the systems, methods and computer program products according to the multiple embodiments of this specification. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of an instruction, and a part of a module, a program segment or an instruction contains one or more executable instructions for realizing the specified logical function. 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 consecutive boxes 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 the flowchart, and the combination of the boxes in the block diagram and / or the flowchart can be implemented by a dedicated hardware-based system that performs the specified function or action, or can be implemented by a combination of dedicated hardware and computer instructions. It is well known to those skilled in the art that it is equivalent to implement it by hardware, implement it by software, and implement it by combining software and hardware.

[0116] The embodiments of the present specification have been described above, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and changes will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The choice of terms used herein is intended to best explain the principles of the embodiments, practical applications, or technical improvements in the marketplace, or to enable other persons of ordinary skill in the art to understand the embodiments disclosed herein. The scope of this application is defined by the appended claims.

Claims

1. A method for identifying user categories, characterized in that: include: Acquire target device information of a terminal device used by a user, and acquire target network environment information of a network environment in which the terminal device is located; Constructing a target graph network representing the association relationship between the user and other users according to the target device information and the target network environment information; Detecting cheating feature information of the user according to the target device information; Performing community analysis on the target graph network using a preset community algorithm to obtain community data corresponding to the user; determining a user vector of the user according to the cheating feature information; calculating a user score of the user according to the community data and the user vector; determining the target user category according to the user score, wherein the user vector includes a plurality of feature elements, and a value of each feature element is determined according to whether the user has corresponding cheating feature information. The step of calculating the user score of the user based on the community data and the user vector includes: obtaining a target weight vector corresponding to the user vector, wherein the target weight vector includes weight elements corresponding to each feature element in the user vector; obtaining the total number of user nodes in the target community to which the user belongs, and obtaining the proportion of user nodes corresponding to the user based on the community data, wherein the proportion of user nodes is a ratio between a first number and a second number of all nodes in the target community, and the first number is the number of all nodes that have a direct connection relationship with the user node of the user; and dividing the product of the user vector, the target weight vector and the proportion of user nodes by the total number of user nodes to obtain the user score.

2. The method according to claim 1, characterized in that The cheating characteristic information includes at least one of first characteristic information indicating whether the terminal device is rooted, second characteristic information indicating whether the terminal device is iOS jailbroken, third characteristic information indicating whether the terminal device is hooked, and fourth characteristic information indicating whether the terminal device is injected.

3. The method according to claim 2, characterized in that The target graph network includes a plurality of graph nodes, and the connection relationship between the plurality of graph nodes is determined according to the device information and network environment information of the corresponding user; The using a preset community algorithm to perform community analysis on the target graph network to obtain community data corresponding to the user includes: According to the connection relationship between the multiple graph nodes in the target graph network, the preset community algorithm is used to perform community analysis to determine the target community to which the user belongs, and data in the target community is obtained as the community data.

4. The method according to claim 1, characterized in that: The method obtains the target weight vector by the following steps: Constructing a first judgment matrix according to the relative importance of all cheating feature information corresponding to the user; Normalizing the first element in each column of the first judgment matrix to obtain a second judgment matrix; Summing the second elements in each row of the second judgment matrix to obtain a third judgment matrix; Obtaining an initial weight vector by calculating the ratio between each third element and the sum of all third elements in the third judgment matrix; The target weight vector is obtained according to the initial weight vector.

5. The method according to claim 4, characterized in that The step of obtaining the target weight vector according to the initial weight vector comprises: When the consistency check is performed on the initial weight vector and the check passes, determining the initial weight vector as the target weight vector; When the consistency check is performed on the initial weight vector and the check fails, the first judgment matrix is ​​reconstructed to determine the target weight vector.

6. A user category identification device, characterized in that: include: An acquisition module, used to acquire target device information of a terminal device used by a user, and to acquire target network environment information of a network environment where the terminal device is located; Identification module for: According to the target device information and the target network environment information, a target graph network representing the association relationship between the user and other users is constructed; according to the target device information, cheating feature information of the user is detected; Performing community analysis on the target graph network using a preset community algorithm to obtain community data corresponding to the user; determining a user vector of the user according to the cheating feature information; calculating a user score of the user according to the community data and the user vector; determining the target user category according to the user score, wherein the user vector includes a plurality of feature elements, and a value of each feature element is determined according to whether the user has corresponding cheating feature information. The step of calculating the user score of the user based on the community data and the user vector includes: obtaining a target weight vector corresponding to the user vector, wherein the target weight vector includes weight elements corresponding to each feature element in the user vector; obtaining the total number of user nodes in the target community to which the user belongs, and obtaining the proportion of user nodes corresponding to the user based on the community data, wherein the proportion of user nodes is a ratio between a first number and a second number of all nodes in the target community, and the first number is the number of all nodes that have a direct connection relationship with the user node of the user; and dividing the product of the user vector, the target weight vector and the proportion of user nodes by the total number of user nodes to obtain the user score.

7. An electronic device, characterized in that: include: A memory for storing executable instructions; A processor is used to operate the electronic device to execute the method according to any one of claims 1 to 5 under the control of the instruction.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program that can be read and executed by a computer, and the computer program is used to execute the method according to any one of claims 1 to 5 when being read and executed by the computer.

Images