User identification method and device, electronic equipment and storage medium
By extracting user behavior information and frequently calculating the risk identification results of the target user, the problem of low payment conversion rate caused by illegal operations is solved, and the recognition accuracy and operational efficiency are improved.
Patent Information
- Application Number
- CN202411033364.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-30
- Publication Date
- 2024-11-12
AI Technical Summary
In the prior art, controlling multiple user accounts frequently participate in incentive tasks through illegal operations, resulting in low payment conversion rate of application operators.
By collecting user behavior information, determining task completion information, performing feature extraction and frequent behavior calculations, risk identification is carried out based on frequent behavior information, and the risk identification results of target users are identified, thereby realizing risk identification of frequent behaviors.
It improves the accuracy of user recognition and reduces the difficulty of recognition, effectively improving the paid conversion rate of application operators.
Smart Images

Figure CN118941334A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of computers, and in particular to a user identification method, device, electronic device and storage medium. Background Art
[0002] With the rapid development of Internet technology, operators of various applications usually adopt various incentive activities to attract users in order to increase user usage. For example, incentive tasks are set up in the application so that users can obtain corresponding reward resources after completing relevant incentive tasks. When the accumulated reward resources reach a certain amount, they can withdraw cash through the wallet.
[0003] Since users are allowed to obtain reward resources by participating in various incentive tasks and eventually withdraw cash, there is no shortage of people who will control multiple user accounts through improper operations to frequently participate in incentive tasks, perform illegal operations to obtain a large amount of reward resources, and cause resource loss to the application operator. Summary of the invention
[0004] In order to solve the problem of low payment conversion rate of application operators caused by controlling multiple user accounts to frequently participate in incentive tasks through illegal operations in the existing related technologies, the present application provides a user identification method, device, electronic device and storage medium.
[0005] In a first aspect, the present application provides a user identification method, comprising:
[0006] Determine task completion information based on the collected user behavior information, wherein the user behavior information is the user's behavior information in the target task link of the application program;
[0007] Extracting features based on the task completion information to obtain user behavior feature information;
[0008] Performing frequent behavior calculation based on the user behavior feature information to obtain frequent behavior information corresponding to the user behavior information;
[0009] Risk identification is performed on a target user corresponding to the frequent behavior information to obtain a risk identification result corresponding to the target user.
[0010] Optionally, determining the task completion information based on the collected user behavior information includes:
[0011] Collecting user behavior information in a target task phase of an application program to obtain the user behavior information;
[0012] Determine a task status corresponding to the user behavior information;
[0013] The task completion information is generated based on the user behavior information that the task status is a task completion status.
[0014] Optionally, performing feature extraction based on the task completion information to obtain user behavior feature information includes:
[0015] Extracting task behavior information and completion time information from the task completion information;
[0016] Feature extraction is performed based on the task behavior information and the completion time information to obtain the user behavior feature information.
[0017] Optionally, the extracting features based on the task behavior information and the completion time information to obtain the user behavior feature information includes:
[0018] Get the preset time interval;
[0019] Determine the completion time information within the time interval as target completion time information;
[0020] Encoding the task behavior information corresponding to the target completion time information to obtain task encoding information corresponding to the time interval;
[0021] The task coding information is used to generate the user behavior feature information.
[0022] Optionally, the performing frequent behavior calculation based on the user behavior feature information to obtain frequent behavior information corresponding to the user behavior information includes:
[0023] Extracting task coding information from the user behavior feature information;
[0024] Performing frequent behavior calculation on the task encoding information to determine the support corresponding to the task encoding information;
[0025] If the support reaches a preset support threshold, the task coding information corresponding to the support is determined as the target coding information;
[0026] The user behavior information corresponding to the target coding information is used to determine the frequent behavior information.
[0027] Optionally, if the support reaches a preset support threshold, determining the task coding information corresponding to the support as the target coding information includes:
[0028] When the support reaches a preset support threshold, extracting encoding parameters from the task encoding information corresponding to the support;
[0029] Determining whether the encoding parameters meet preset encoding conditions;
[0030] When the encoding parameters meet the encoding conditions, the task encoding information is determined as the target encoding information.
[0031] Optionally, the performing risk identification on the target user corresponding to the frequent behavior information to obtain a risk identification result corresponding to the target user includes:
[0032] Determine the user corresponding to the frequent behavior information as the target user;
[0033] Perform risk identification on the target users to obtain a percentage of risky users;
[0034] Obtain the user management and control strategy corresponding to the analyzed user ratio value;
[0035] The user management and control strategy is adopted to generate a risk identification result corresponding to the target user.
[0036] In a second aspect, the present application provides a user identification device, including:
[0037] A determination module, used to determine task completion information based on the collected user behavior information, wherein the user behavior information is the behavior information of the user in the target task link of the application program;
[0038] A feature extraction module, used to extract features based on the task completion information to obtain user behavior feature information;
[0039] A calculation module, used to calculate frequent behaviors based on the user behavior feature information to obtain frequent behavior information corresponding to the user behavior information;
[0040] The risk identification module is used to perform risk identification on the target user corresponding to the frequent behavior information to obtain a risk identification result corresponding to the target user.
[0041] In a third aspect, an electronic device is provided, comprising a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus;
[0042] Memory, used to store computer programs;
[0043] The processor is used to implement the user identification method described in any one of the first aspects when executing the program stored in the memory.
[0044] According to a fourth aspect, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the user identification method as described in any one of the first aspects is implemented.
[0045] The embodiment of the present application determines task completion information based on collected user behavior information, where the user behavior information is the behavior information of the user in the target task link of the application, and performs feature extraction based on the task completion information to obtain user behavior feature information, and performs frequent behavior calculation based on the user behavior feature information to obtain frequent behavior information corresponding to the user behavior information, and then performs risk identification based on the target user corresponding to the frequent behavior information to obtain a risk identification result corresponding to the target user; thereby achieving the purpose of risk identification for frequent behaviors, solving the problem of low payment conversion rate of application operators caused by controlling multiple user accounts to frequently participate in incentive tasks through illegal operations in the existing related technologies, and effectively improving the payment conversion rate. BRIEF DESCRIPTION OF THE DRAWINGS
[0046] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.
[0047] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the existing related technologies, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.
[0048] Figure 1 A flowchart of a user identification method provided in an embodiment of the present application;
[0049] Figure 2 A schematic diagram of the structure of a user identification device provided in an embodiment of the present application;
[0050] Figure 3 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION
[0051] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.
[0052] In order to increase the activity of users in the application, the operator of the application will set up one or more incentive tasks for users to participate, such as signing in, watching time, shopping, downloading other applications, opening treasure chests, etc. Users can obtain reward resources by completing relevant incentive tasks. The reward resources can be gold coins, points and other resources, and users can withdraw cash through the reward resources, which can play a role in increasing user activity.
[0053] Since users are allowed to obtain reward resources by participating in various activities and eventually withdraw cash, there will be risky people who frequently participate in incentive tasks through illegal means. For example, one person controls multiple devices or multiple user accounts, performs illegal operations and frequently participates in incentive tasks, collecting a large amount of reward resources and causing losses to the application operator.
[0054] In order to prevent the application's reward resources from being illegally obtained by risky personnel, existing related technologies will identify users to determine whether they are risky users, thereby managing and restricting risky users, thereby preventing reward resources from being obtained in large quantities by risky users; the user identification method is usually to analyze the behavior of each user, and when the user's behavior shows preset risky behavior, the user's behavior is identified as a risky user. Although this identification method can identify users, it can only identify individual users through the user's own behavior data, and as user activities increase, the difficulty of managing and identifying user behavior also increases, resulting in difficulty in user identification.
[0055] In order to solve the problem of low payment conversion rate of application operators caused by controlling multiple user accounts to frequently participate in incentive tasks through illegal operations in the existing related technologies, the present application provides a user identification method, device, electronic device and storage medium, by determining task completion information based on collected user behavior information, the user behavior information is the behavior information of the user in the target task link of the application, and performing feature extraction based on the task completion information to obtain user behavior feature information, and performing frequent behavior calculation based on the user behavior feature information to obtain frequent behavior information corresponding to the user behavior information, and then performing risk identification based on the target user corresponding to the frequent behavior information to obtain the risk identification result corresponding to the target user; thereby achieving the purpose of risk identification of frequent behaviors, that is, it is no longer limited to identification of a single user behavior data, but is analyzed and identified based on the frequent behavior information corresponding to the user behavior information, which plays a role in improving identification accuracy and reducing identification difficulty; and it also solves the problem of low payment conversion rate of application operators caused by controlling multiple user accounts to frequently participate in incentive tasks through illegal operations in the existing related technologies, and effectively improves the payment conversion rate.
[0056] Figure 1A flow chart of a user identification method provided for an embodiment of the present application. The method can be applied to one or more electronic devices such as an application, a client, and a server. In addition, the execution subject of the method can be hardware or software. When the above-mentioned execution subject is hardware, the execution subject can be one or more of the above-mentioned electronic devices. For example, a single electronic device can execute the method, or a plurality of electronic devices can cooperate with each other to execute the method. When the above-mentioned execution subject is software, the method can be implemented as multiple software or software modules, or as a single software or software module. No specific limitation is made here.
[0057] like Figure 1 As shown, the present application discloses an embodiment, providing a user identification method, which may specifically include the following steps:
[0058] Step S110: Determine task completion information based on the collected user behavior information, where the user behavior information is the user's behavior information in the target task link of the application.
[0059] Specifically, the user's behavior information can be collected after the application is started. The user behavior information can include user behavior and behavior time, etc., wherein the user behavior information corresponding to all users in the same application can be collected, and the user behavior information of related applications can also be obtained as needed, such as computer-side applications and mobile-side applications, etc., which are not specifically limited in this embodiment; it should be noted that the user behavior information in this embodiment can be the user's behavior information in the target task link of the application, wherein the target task link represents the task link pre-configured in the application, such as the sign-in task link, the video watching task link, the treasure chest opening task link, etc., that is, the target task link represents the task link that currently needs to be detected, such as for sending The task link for releasing reward resources can avoid using illegal operations to obtain a large amount of reward resources in the target task link, and the user behavior information in this embodiment can be the behavior entered by the user in the target task link; then the task completion information can be generated based on the user behavior information, and the task completion information can indicate the user behavior information that the user has completed the relevant task in the target task link. Since the target task link may contain a large number of invalid behaviors, and invalid behaviors will not trigger the issuance of reward resources, the task completion information at this time indicates the user's valid behavior in the target task link, which can trigger the issuance of reward resources, thereby filtering the user behavior information and only retaining the task completion information that can trigger resource rewards for the user in the target task link.
[0060] In one example, taking the target task link as a treasure chest opening task environment as an example, in the treasure chest opening task link, the user's viewing description behavior, exit behavior, and other behaviors that cannot trigger the issuance of rewards are invalid behaviors, while the user's clicking on the treasure chest behavior, opening the treasure chest behavior, and other behaviors that can trigger the issuance of reward resources are valid behaviors. At this time, the user behavior information corresponding to the valid behavior can be determined as the task completion information; of course, the above is only an example description, and adaptive adjustments can be made according to needs and scenarios during specific implementation, and this embodiment does not make specific limitations on this.
[0061] Step S120: Extract features based on the task completion information to obtain user behavior feature information.
[0062] Specifically, after determining the task completion information, feature extraction can be performed on the task completion information to obtain user behavior feature information, wherein feature extraction means extracting user behavior features from the task completion information, and the user behavior features may include behavior identifiers, behavior time and other information. The behavior identifier may be used to mark specific user behaviors, and the behavior time is used to indicate the time when the user behavior corresponding to the task completion information is executed; thus, the user behavior feature information may indicate information including the association between user behavior and time.
[0063] Step S130: Calculate frequent behaviors based on the user behavior feature information to obtain frequent behavior information corresponding to the user behavior information.
[0064] Specifically, since the features of each task completion information can be extracted to obtain the user behavior feature information, and each task completion information corresponds to the user behavior information, that is, the user behavior feature information is mutually corresponding to the user behavior information; after obtaining the user behavior feature information, the user behavior feature information can be used to perform frequent behavior calculation to obtain the frequent behavior information corresponding to the user behavior information, wherein the frequent behavior calculation means calculating the frequent times of each user behavior feature based on all the user behavior feature information, and determining the user behavior corresponding to the user behavior feature whose frequent times reaches a preset frequent times threshold as the frequent behavior information. The specific frequent behavior calculation method can adopt a neural network model, such as a PrefixSpan model; thereby playing the role of calculating the frequent behavior information from the user behavior information, so that the frequent behavior information can represent the behavior information that frequently appears in all the user behavior feature information.
[0065] It should be noted that in this embodiment, the frequent behavior information may include one or more user behavior features, and the user behavior features may include behavior identifiers, behavior time and other information, that is, the frequent behavior information may include one or more behavior identifiers, behavior time and other information.
[0066] In one example, the user behavior feature information is [A, B], [C, D], where different [ ] represent different behavior times, and A, B, C, D represent specific user behaviors. The frequent behavior information obtained at this time can be [A, B], [C, D], or [A, B], [C, D]. Of course, the above is only an example for illustration. During specific implementation, adaptive adjustments can be made according to needs and scenarios, and this embodiment does not make specific limitations on this.
[0067] Step S140: Perform risk identification on the target user corresponding to the frequent behavior information to obtain a risk identification result corresponding to the target user.
[0068] Specifically, after obtaining frequent behavior information, since frequent behavior information can represent behavior information that frequently appears in all user behavior feature information, the user indicated by the frequent behavior information can be regarded as the target user, and risk identification is performed on the target user to obtain a risk identification result corresponding to the target user, wherein the risk identification method can be to use a neural network model to identify and analyze the user information corresponding to the target user, or to identify in combination with a preset risk user database, which is not specifically limited in this embodiment; and the risk identification result obtained can indicate whether the target user is a risky user, and can also indicate the corresponding management and control method of the target user, etc., which is not specifically limited in this embodiment.
[0069] In one example, when the frequent behavior information is [A, B], [C, D], users with behaviors corresponding to [A, B], [C, D] can be identified as target users, and then each target user is judged to be a risky user according to a preset risk user database. The risk identification result obtained at this time can be the result that each target user is a risky user or not a risky user.
[0070] This embodiment determines task completion information based on the collected user behavior information, where the user behavior information is the user's behavior information in the target task link of the application, and performs feature extraction based on the task completion information to obtain user behavior feature information, and performs frequent behavior calculation based on the user behavior feature information to obtain frequent behavior information corresponding to the user behavior information, and then performs risk identification based on the target user corresponding to the frequent behavior information to obtain a risk identification result corresponding to the target user; that is, by analyzing and processing all user behavior information in the application, frequently occurring frequent behavior information is obtained, so that the user corresponding to the frequent behavior information is determined as the target user, and then risk identification is performed for each target user, which plays a role in risk identification of users related to frequent behaviors, and can avoid the problem of low payment conversion rate of the application operator caused by illegal operations to control multiple user accounts to frequently participate in incentive tasks. By performing risk identification on users related to frequent behaviors, the payment conversion rate can be effectively improved.
[0071] In an optional embodiment of the present application, step S110 determines the task completion information based on the collected user behavior information, and may specifically include the following sub-steps: collecting the user's behavior information in the target task link of the application to obtain user behavior information; determining the task status corresponding to the user behavior information; and generating task completion information based on the user behavior information whose task status is the task completion status.
[0072] In this embodiment, in the process of determining the task completion information, the user's behavior information in the target task link of the application program can be collected. The behavior information can be any behavior of the user in the target task link, such as clicking, viewing, exiting, etc., so as to obtain the user behavior information, and then the task status corresponding to the user behavior information can be determined. Since the user behavior information is the user's behavior information in the target task link of the application program, different user behaviors will obtain different task statuses. The task status indicates the status of the task corresponding to the current user behavior information. For example, in the treasure chest opening task link, the user's opening treasure chest behavior will obtain the task completion status, while the user's clicking to view the rules behavior will obtain the task incomplete status. After determining the task status, it can be determined whether the task status is the task completion status. If the task status is the task completion status, the user behavior information whose task status is the task completion status can be determined as the task completion information, thereby filtering the user behavior information and only filtering out the user behavior that has completed the relevant tasks in the target task link.
[0073] In an optional embodiment of the present application, step S120 performs feature extraction based on task completion information to obtain user behavior feature information, which may specifically include the following sub-steps: extracting task behavior information and completion time information from the task completion information; performing feature extraction based on the task behavior information and completion time information to obtain user behavior feature information.
[0074] In this embodiment, after obtaining the task completion information, task behavior information and completion time information can be extracted from the task completion information, the task behavior information represents the behavior input by the user in the target task link, and the completion time information represents the task completion time in the target task link, and then feature extraction can be performed based on the task behavior information and completion time information to obtain user behavior feature information, wherein a specific feature extraction method can be to extract a task identifier from the task behavior information, extract the completion time from the completion time information, and generate user behavior feature information based on the task identifier and the completion time, so that the user behavior feature information can represent information including the association between user behavior and time for completing related tasks.
[0075] In an optional embodiment of the present application, feature extraction is performed based on task behavior information and completion time information to obtain user behavior feature information, which may specifically include the following sub-steps: obtaining a preset time interval; determining the completion time information within the time interval as the target completion time information; encoding the task behavior information corresponding to the target completion time information to obtain task coding information corresponding to the time interval; and using the task coding information to generate user behavior feature information.
[0076] In this embodiment, in the process of obtaining user behavior characteristic information, a preset time interval can be obtained. The preset time interval represents a pre-configured time area, which can be 1 hour in 24 hours a day configured as a time interval, or 0.5 hours configured as an interval, etc. This embodiment does not make specific limitations on this; thereby, multiple time intervals without time intersection and continuous can be obtained, and then for each time area, the completion time information belonging to the time interval can be determined as the target completion time information, so that the target completion time information can contain the association relationship information between the time interval and the completion time; and then the task behavior information corresponding to the target completion time information in the same time area can be encoded and processed to obtain the task coding information corresponding to the time interval, and the task coding information can represent the information of the association relationship between the time area and the task behavior information, and the coding processing can be to encode and associate the time interval corresponding to the target time information with the task behavior information; then the task coding information can be determined as the user behavior characteristic information.
[0077] In one example, each of the 24 hours in a day can be divided into a time interval, and then 24 time intervals without time intersection and continuous can be obtained, and for each time zone, the completion time information belonging to the time interval is determined as the target completion time information. For example, the completion time information belonging to the time zone of 9-10 o'clock includes 9:10, 9:30, and 9:59. Since 9:10, 9:30, and 9:59 all belong to the time zone of 9-10 o'clock, the completion time information corresponding to 9:10, 9:30, and 9:59 can be determined as the target completion time information, and then the target completion time information can be determined for 9:10, 9:30, and 9:59. The task behavior information corresponding to the time information is encoded and processed, and the task coding information corresponding to the time interval is [open treasure chest*4, sign in*1, watch video*30]. At this time, [] represents the time zone from 9 to 10 o'clock, where "open treasure chest*4, sign in*1, watch video*30" is the task behavior information corresponding to the target completion time information, which means that the user completed the task of opening treasure chest 4 times, signing in once, and watching video 30 times in the time zone from 9 to 10 o'clock. At this time, the task coding information can be determined as user behavior feature information. It should be noted that the tasks in the same time interval do not need to consider the order. If there is no completed task in a certain time interval, an empty code [] can be obtained.
[0078] In addition, the time interval can also contain multiple time levels. For example, the task coding information corresponding to the 24 time intervals of 24 hours in a day can be [[...]...[watch video*30, sign in*1]...[watch video*30]...[open treasure chest*4]...[...]], where [ ] can represent the time area corresponding to a certain day, and each [ ] in [ ] represents the time area corresponding to an hour, and finally the task coding information of each user for 24 hours is obtained.
[0079] In an optional embodiment of the present application, step S130 performs frequent behavior calculations based on user behavior feature information to obtain frequent behavior information corresponding to the user behavior information, which may specifically include the following sub-steps: extracting task coding information from the user behavior feature information; performing frequent behavior calculations on the task coding information to determine the support corresponding to the task coding information; if the support reaches a preset support threshold, determining the task coding information corresponding to the support as the target coding information; and determining the frequent behavior information from the user behavior information corresponding to the target coding information.
[0080] In this embodiment, after obtaining the user behavior feature information, task coding information can be extracted from the user behavior feature information, and frequent behavior calculation can be performed on the task coding information to determine the support corresponding to the task coding information, wherein the frequent behavior calculation can be performed using a neural network model to determine the support corresponding to the task coding information, and the support represents the frequency of a sequence appearing in a data set, wherein the data set can represent a set of any one or more task coding information corresponding to each user, and the sequence can represent a set of all task coding information corresponding to all users, so that the support can represent the frequency of any one or more task coding information corresponding to a certain user in the set of all task coding information corresponding to all users; then it can be determined whether the support reaches a preset support threshold, and the preset support threshold represents a pre-configured support threshold that meets the frequent behavior. When the support reaches the preset support threshold, it can be determined that the frequency of occurrence of the current task coding information meets the preset frequent behavior frequency. At this time, the task coding information corresponding to the support can be determined as the target coding information, and the frequent behavior information is generated based on the user behavior information corresponding to the target coding information. The specific generation method can be to sort the user behavior information in chronological order, which is not specifically limited in this embodiment.
[0081] It should be noted that the frequent behavior information in this embodiment may include one or more target task coding information, and there is a time correlation relationship between each target task coding information, and the target coding information will have corresponding user information. At this time, the target coding information can be sorted in chronological order. For example, the target coding information includes [watch video*30, sign in*1] corresponding to the first hour interval, [watch video*30] corresponding to the second hour interval, and [open treasure chest*4] corresponding to the third hour interval. At this time, the frequent behavior information obtained by sorting according to time is [[watch video*30, sign in*1], [watch video*30], [open treasure chest*4]], which means that 30 video watching activities and 1 sign in activity were completed in the first hour, 30 video watching activities were completed in the second hour, and 4 treasure chest opening activities were completed in the third hour. It should be noted that the first hour here The first, second and third hours do not need to be consecutive, as long as the first hour is earlier than the second and third hours in time sequence. Secondly, the task coding information corresponding to the time interval only needs to be the user's actual user behavior information within the time interval. For example, the first hour in the aforementioned example represents the completion of 30 video watching activities and 1 sign-in activity. At this time, a user completed 30 video watching activities, 1 sign-in activity, 1 treasure chest opening, and 2 pet interactions at 8 o'clock in the morning. The task coding information corresponding to the time interval can also be matched with [watch video*30, sign-in*1]. That is, for [watch video*30, sign-in*1] in the frequent behavior information, it is sufficient to include 30 video activities and 1 sign-in activity in a time interval. The coding matching in other frequent behavior information is similar to this and will not be repeated here.
[0082] In one example, the PrefixSpan model provided by the spark framework can be used to calculate the task encoding information. In this case, it is necessary to set the corresponding preset support threshold for the PrefixSpan model in advance. The preset support threshold is the minimum support of the PrefixSpan model. In the PrefixSpan model for association rule mining, the support is used to evaluate the coverage of frequent behavior sequences, so that the PrefixSpan model can output the target encoding information with support reaching the preset support threshold.
[0083] In an optional embodiment of the present application, if the support reaches a preset support threshold, the task coding information corresponding to the support is determined as the target coding information, which may specifically include the following sub-steps: when the support reaches a preset support threshold, extracting coding parameters from the task coding information corresponding to the support; determining whether the coding parameters meet preset coding conditions; when the coding parameters meet the coding conditions, determining the task coding information as the target coding information.
[0084] In this embodiment, when the support reaches the preset support threshold, encoding parameters can be extracted from the task encoding information corresponding to the support, and the encoding parameters represent the parameters of the data contained in the task encoding information, such as the number of task types, the number of time intervals, the number of user behaviors, and other parameters contained in the task encoding information; then it can be determined whether the encoding parameters meet the preset encoding conditions, and the preset encoding conditions represent the conditions configured for each encoding parameter in advance according to the needs, such as the number of types that the number of task types needs to reach, the number of time intervals that the number of intervals needs to reach, and the number of user behaviors that the number of behaviors needs to reach. Therefore, when the encoding parameters do not meet the encoding conditions, there is no need to perform subsequent steps, and when the encoding parameters meet the encoding conditions, the task encoding information can be determined as the target encoding information. This prevents the task encoding information from being too simple and misidentifying normal users.
[0085] In one example, when the encoding parameter is the number of task types, the encoding condition is the type number threshold that the number of task types needs to reach, and the type number threshold is 2, if the task coding information is [[watch video*30], [watch video*28]], the extracted coding parameter from the task coding information is the number of types "1", it can be determined that the coding parameter does not meet the preset coding condition, and there is no need to perform subsequent steps. Since watching videos itself is the most common behavior of users, if the task coding information is used to identify risky users, it will easily injure normal users, so they can be screened out by the coding condition.
[0086] When the encoding parameter is the number of time intervals, the encoding condition is the interval number threshold that the number of time intervals needs to reach, and the interval number threshold is 5, if the task coding information is [[watch video *31, sign in *1]], the extracted coding parameter from the task coding information is the interval number threshold "2", it can be determined that the coding parameter does not meet the preset coding condition, and there is no need to perform subsequent steps. Since the task coding information only contains behavior for 1 hour, if the task coding information is used to identify risky users, it will be easier to accidentally injure normal users, so they can be screened out by the coding condition.
[0087] When the encoding parameter is the number of user behaviors, the encoding condition is the threshold of the number of user behaviors that must be reached, and the threshold of the number of behaviors is 30, if the task encoding information is [[watch videos*15, sign in*1], [open treasure chest*1]], the encoding parameter extracted from the task encoding information is the number of user behaviors "17", it can be determined that the encoding parameter does not meet the preset encoding condition, and there is no need to perform subsequent steps. Since there are a total of 17 user behaviors in the task encoding information, which is a case of a small number of behaviors, if the task encoding information is used to identify risky users, it will be easier to accidentally injure normal users, so they can be screened out by the encoding condition.
[0088] In an optional embodiment of the present application, step S140 performs risk identification based on the target user corresponding to the frequent behavior information to obtain a risk identification result corresponding to the target user, which may specifically include the following sub-steps: determining the user corresponding to the frequent behavior information as the target user; performing risk identification on the target user to obtain a risk user ratio value; obtaining a user management strategy corresponding to the analysis user ratio value; and adopting the user management strategy to generate a risk identification result corresponding to the target user.
[0089] In this embodiment, after obtaining the frequent behavior information, the user corresponding to the frequent behavior information can be determined as the target user. Since the frequent behavior information can contain multiple target task coding information sorted by time, the user corresponding to the user behavior information that meets the frequent behavior information is determined as the target user, so that risk identification can be performed on the target user to obtain the risk user ratio value, wherein risk identification means comparing the target user with a preset risk user database, so as to determine whether each target user belongs to a risk user in the risk user database, so that the risk user ratio value can represent the ratio value of risk users among the target users, and the risk user database can be a database generated by other identification strategies of the application, or a database pre-configured according to requirements; then the user control strategy corresponding to the analysis user ratio value can be obtained, since different user ratio values indicate different numbers of risk users among the target users, different user control strategies can be used to control the target users, wherein the user control strategy represents a strategy for controlling each target user, and then the user control strategy can be used to control each target user, and the risk identification result corresponding to the target user is generated, and the risk identification result can indicate whether the target user is a risk user and the control strategy corresponding to the target user.
[0090] In one example, if the frequent behavior information is [[watch videos*30, sign in*1], [watch videos*30], [open treasure chests*4]], then the target task coding information corresponding to each user can be retrieved based on the frequent behavior information, and the user corresponding to the user behavior information who completed 30 video watching activities and 1 sign in activity in the first hour, 30 video watching activities in the second hour, and 4 treasure chest opening activities in the third hour is determined as the target user; then, when the target users determined based on the frequent behavior information [[watch videos*30, sign in*1], [watch videos*30], [open treasure chests*4]] are 1,000, it can be determined to perform risk identification on the target users and obtain the percentage of risky users. For example, 900 of the 1,000 target users are at risk. There are high-risk users in the destiny, and the proportion of risk users is 90% at this time. Therefore, the user control strategy corresponding to the 90% risk user proportion is to intercept and control 900 risk users, and handle the remaining 100 non-high-risk users offline; if another frequent behavior information [[watch video*31, sign in*1], [watch video*15], [open treasure chest*3]] also hits 1,000 target users, of which 700 users hit high risk on the same day, and the proportion of risk users is 70% at this time, it is determined that the user control strategy corresponding to the 70% risk user proportion is to superimpose other control strategies of the application for control. For example, the remaining non-risk users also hit the device abnormality strategy. At this time, the non-risk user control can be combined with the device abnormality control strategy.
[0091] like Figure 2 As shown, the present application also discloses an embodiment, providing a user identification device, which may include the following modules:
[0092] A determination module 210 is used to determine task completion information based on the collected user behavior information, where the user behavior information is the behavior information of the user in the target task link of the application program;
[0093] A feature extraction module 220 is used to extract features based on the task completion information to obtain user behavior feature information;
[0094] A calculation module 230, configured to perform frequent behavior calculation based on the user behavior feature information to obtain frequent behavior information corresponding to the user behavior information;
[0095] The risk identification module 240 is used to perform risk identification on the target user corresponding to the frequent behavior information to obtain a risk identification result corresponding to the target user.
[0096] In an optional embodiment of the present application, the determination module 210 may include:
[0097] A collection unit, used to collect user behavior information in a target task link of an application program to obtain the user behavior information;
[0098] A first determining unit, used to determine a task state corresponding to the user behavior information;
[0099] The first generating unit is used to generate the task completion information based on the user behavior information that the task status is a task completion status.
[0100] In an optional embodiment of the present application, the feature extraction module 220 may include:
[0101] A first extraction unit, used to extract task behavior information and completion time information from the task completion information;
[0102] The second extraction unit is used to perform feature extraction based on the task behavior information and the completion time information to obtain the user behavior feature information.
[0103] In an optional embodiment of the present application, the second extraction unit may include:
[0104] A first acquisition subunit, used to acquire a preset time interval;
[0105] A first determining subunit is used to determine the completion time information belonging to the time interval as the target completion time information;
[0106] A coding processing subunit, used for coding the task behavior information corresponding to the target completion time information to obtain the task coding information corresponding to the time interval;
[0107] The first generating subunit is used to generate the user behavior feature information by using the task coding information.
[0108] In an optional embodiment of the present application, the calculation module 230 may include:
[0109] A third extraction unit, used to extract task coding information from the user behavior feature information;
[0110] A calculation unit, used to perform frequent behavior calculation on the task encoding information to determine the support corresponding to the task encoding information;
[0111] A second determining unit is used to determine the task coding information corresponding to the support as the target coding information if the support reaches a preset support threshold;
[0112] The third determining unit is used to determine the frequent behavior information by using the user behavior information corresponding to the target coding information.
[0113] In an optional embodiment of the present application, the second determining unit may include:
[0114] An extraction subunit, configured to extract encoding parameters from the task encoding information corresponding to the support when the support reaches a preset support threshold;
[0115] A second determining subunit, used to determine whether the encoding parameters meet the preset encoding conditions;
[0116] The third determining subunit is used to determine the task encoding information as target encoding information when the encoding parameter meets the encoding condition.
[0117] In an optional embodiment of the present application, the risk identification module 240 may include:
[0118] A fourth determining unit, configured to determine the user corresponding to the frequent behavior information as the target user;
[0119] A risk identification unit, used to identify risks for the target users and obtain a percentage value of risky users;
[0120] An acquisition unit, used to acquire a user management and control strategy corresponding to the analyzed user proportion value;
[0121] The second generating unit is used to adopt the user management and control strategy to generate a risk identification result corresponding to the target user.
[0122] The implementation process of the functions and effects of each module in the above-mentioned device is specifically described in the implementation process of the corresponding steps in the above-mentioned method, which will not be repeated here.
[0123] like Figure 3 As shown, an embodiment of the present application provides an electronic device, including a processor 310, a communication interface 320, a memory 330 and a communication bus 340, wherein the processor 310, the communication interface 320, and the memory 330 communicate with each other through the communication bus 340;
[0124] Memory 330, for storing computer programs;
[0125] In one embodiment of the present application, the processor 310, when executing the program stored in the memory 330, implements the user identification method provided by any of the aforementioned method embodiments, by determining task completion information based on the collected user behavior information, the user behavior information being the user's behavior information in the target task link of the application, and performing feature extraction based on the task completion information to obtain user behavior feature information, and performing frequent behavior calculation based on the user behavior feature information to obtain frequent behavior information corresponding to the user behavior information, and then performing risk identification based on the target user corresponding to the frequent behavior information to obtain a risk identification result corresponding to the target user; thereby achieving the purpose of risk identification of frequent behaviors, that is, it is no longer limited to identification of a single user behavior data, but is analyzed and identified based on the frequent behavior information corresponding to the user behavior information, which improves the accuracy of identification and reduces the difficulty of identification; and also solves the problem of low payment conversion rate of the application operator caused by controlling multiple user accounts to frequently participate in incentive tasks through illegal operations in the existing related technologies, thereby effectively improving the payment conversion rate.
[0126] The embodiment of the present application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements a user identification method as provided in any of the aforementioned method embodiments, by determining task completion information based on collected user behavior information, the user behavior information being the behavior information of the user in the target task link of the application, and performing feature extraction based on the task completion information to obtain user behavior feature information, and performing frequent behavior calculation based on the user behavior feature information to obtain frequent behavior information corresponding to the user behavior information, and then performing risk identification based on the target user corresponding to the frequent behavior information to obtain a risk identification result corresponding to the target user; thereby achieving the purpose of risk identification of frequent behaviors, that is, no longer limited to identification of a single user behavior data, but rather analyzing and identifying with the frequent behavior information corresponding to the user behavior information, which improves identification accuracy and reduces identification difficulty; and also solves the problem of low payment conversion rate of application operators caused by controlling multiple user accounts to frequently participate in incentive tasks through illegal operations in the existing related technologies, thereby effectively improving payment conversion rate.
[0127] It should be noted that, in this article, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.
[0128] The above describes specific embodiments of the present specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
[0129] The foregoing is merely a specific embodiment of the present invention, which enables those skilled in the art to understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features claimed herein.
Claims
1. A user identification method, characterized in that: include: Determine task completion information based on the collected user behavior information, wherein the user behavior information is the user's behavior information in the target task link of the application program; Extracting features based on the task completion information to obtain user behavior feature information; Performing frequent behavior calculation based on the user behavior feature information to obtain frequent behavior information corresponding to the user behavior information; Risk identification is performed on a target user corresponding to the frequent behavior information to obtain a risk identification result corresponding to the target user.
2. The user identification method according to claim 1, characterized in that: The task completion information is determined based on the collected user behavior information, including: Collecting user behavior information in a target task phase of an application program to obtain the user behavior information; Determine a task status corresponding to the user behavior information; The task completion information is generated based on the user behavior information that the task status is a task completion status.
3. The user identification method according to claim 1, characterized in that: The feature extraction is performed based on the task completion information to obtain user behavior feature information, including: Extracting task behavior information and completion time information from the task completion information; Feature extraction is performed based on the task behavior information and the completion time information to obtain the user behavior feature information.
4. The user identification method according to claim 3, characterized in that: The extracting features based on the task behavior information and the completion time information to obtain the user behavior feature information includes: Get the preset time interval; Determine the completion time information within the time interval as target completion time information; Encoding the task behavior information corresponding to the target completion time information to obtain task encoding information corresponding to the time interval; The task coding information is used to generate the user behavior feature information.
5. The user identification method according to claim 1, characterized in that: The performing frequent behavior calculation based on the user behavior feature information to obtain frequent behavior information corresponding to the user behavior information includes: Extracting task coding information from the user behavior feature information; Performing frequent behavior calculation on the task encoding information to determine the support corresponding to the task encoding information; If the support reaches a preset support threshold, the task coding information corresponding to the support is determined as the target coding information; The user behavior information corresponding to the target coding information is used to determine the frequent behavior information.
6. The user identification method according to claim 5, characterized in that: If the support reaches a preset support threshold, determining the task coding information corresponding to the support as the target coding information includes: When the support reaches a preset support threshold, extracting encoding parameters from the task encoding information corresponding to the support; Determining whether the encoding parameters meet preset encoding conditions; When the encoding parameters meet the encoding conditions, the task encoding information is determined as the target encoding information.
7. The user identification method according to claim 1, characterized in that: The performing risk identification on the target user corresponding to the frequent behavior information to obtain the risk identification result corresponding to the target user includes: Determine the user corresponding to the frequent behavior information as the target user; Perform risk identification on the target users to obtain a percentage of risky users; Obtain the user management and control strategy corresponding to the analyzed user ratio value; The user management and control strategy is adopted to generate a risk identification result corresponding to the target user.
8. A user identification device, characterized in that: include: A determination module, used to determine task completion information based on the collected user behavior information, wherein the user behavior information is the behavior information of the user in the target task link of the application program; A feature extraction module, used to extract features based on the task completion information to obtain user behavior feature information; A calculation module, used to calculate frequent behaviors based on the user behavior feature information to obtain frequent behavior information corresponding to the user behavior information; The risk identification module is used to perform risk identification on the target user corresponding to the frequent behavior information to obtain a risk identification result corresponding to the target user.
9. An electronic device, characterized in that: It includes a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other through the communication bus; Memory, used to store computer programs; The processor is used to implement the user identification method described in any one of claims 1 to 7 when executing the program stored in the memory.
10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the user identification method according to any one of claims 1 to 7 is implemented.
