A method and device for determining account grouping, electronic equipment and storage medium

Abstract

The present disclosure relates to a method and device for determining account groups, electronic equipment and storage medium. The method for determining account groups comprises: randomly selecting a preset number of sub-buckets from a bucket set to obtain a candidate account group, each sub-bucket comprising a plurality of accounts; obtaining first behavior data of accounts in the candidate account group and second behavior data of accounts in the remaining sub-buckets, the remaining sub-buckets being sub-buckets other than the candidate account group in the bucket set; determining decomposition features of the first behavior data and the second behavior data in a preset time period, respectively; and determining a plurality of first account groups from a plurality of candidate account groups according to the decomposition features. In the present disclosure, the selected account groups have the technical effect of having minimal differences, and when these account groups are used for subsequent scheme comparison tests, the differences between the account groups are minimal, ensuring the accuracy of the scheme comparison test.

Description

Technical Field

[0001] This disclosure relates to the field of computer network technology, and in particular to a method, apparatus, electronic device and storage medium for determining account groups. Background Technology

[0002] With the rapid development of computer network technology, software products are also iterating very quickly. During product iteration, determining whether an updated solution is more optimized than a previous one, or which of several updated solutions is the best, can be achieved through the following method: For example, develop two or more solutions for the same product. For instance, two different solutions can be developed for the same tab page. Then, randomly select a portion of user terminals to execute the first solution and another portion to execute the second solution. By comparing relevant metrics of user data from the first and second solutions, it can be predicted which solution best meets the requirements.

[0003] In related technologies, before comparing solutions, refer to Figure 1 The method uses a hash algorithm to randomly aggregate and bucket all user accounts, and then randomly groups these buckets together. However, due to random sampling, this grouping method has an inherent error, and there is a significant probability that the groups will differ substantially from each other. When comparing different schemes using these significantly different groups, the resulting data is inaccurate, leading to inaccurate scheme comparisons.

[0004] Therefore, there is an urgent need for a method to determine account grouping in order to improve the accuracy of scheme comparison. Summary of the Invention

[0005] This disclosure provides a method, apparatus, electronic device, and storage medium for determining account groups, to at least solve the problem of inaccurate comparison of solutions in related technologies. The technical solution of this disclosure is as follows:

[0006] According to a first aspect of the present disclosure, a method for determining account groups is provided, comprising:

[0007] A preset number of buckets are randomly selected from the bucket set to obtain candidate account groups. Each bucket includes multiple accounts. The bucket set is obtained by bucketing accounts based on their identification information.

[0008] Obtain the first row data of the accounts in the candidate account group and the second row data of the accounts in the remaining buckets, wherein the remaining buckets are the buckets in the bucket set other than the candidate account group;

[0009] Determine the decomposition features of the first behavioral data and the second behavioral data within a preset time period, respectively;

[0010] Multiple first account groups are determined from multiple candidate account groups based on the decomposition features.

[0011] In one possible implementation, obtaining the first behavioral data of accounts in the candidate account group and the second behavioral data of accounts in the remaining buckets includes:

[0012] Obtain the mean and variance of the behavioral data of accounts in the bucket set over a preset time period;

[0013] The behavioral data is standardized based on the mean and variance to obtain standard behavioral data.

[0014] Obtain the first behavioral data of the accounts in the candidate account group and the second behavioral data of the accounts in the remaining buckets from the standard behavioral data.

[0015] In one possible implementation, determining multiple first account groups from multiple candidate account groups based on the decomposition features includes:

[0016] Obtain the difference between the decomposition features of the first behavioral data within a preset time period and the decomposition features of the second behavioral data within a preset time period, and determine a preset number of candidate account groups with the smallest difference;

[0017] Obtain the sliding average of the first row data of each candidate account group within a preset period, and group the candidate account group with the smallest sliding average as a first account group;

[0018] Remove the buckets of the account groups from the bucket set to update the bucket set;

[0019] Based on the updated bucket set, the following steps are repeated: obtaining the difference between the decomposition features of the first and second behavioral data within a preset time period, determining candidate account groups, obtaining the sliding average of the first behavioral data of the candidate accounts within a preset period, and obtaining the next first account group, until the number of first account groups obtained reaches a preset value.

[0020] In one possible implementation, determining multiple first account groups from multiple candidate account groups based on the decomposition features includes:

[0021] Multiple initial first account groups are determined from multiple candidate account groups based on the decomposition features;

[0022] Determine the significance of the difference between any two initial first account groups among the plurality of initial first account groups. If no two initial first account groups are significant, then the plurality of initial first account groups are taken as the first account group.

[0023] In one possible implementation, the method further includes:

[0024] Obtain multiple account groups, and select one first account group from the multiple account groups as a second account group, wherein each account group includes the multiple first account groups;

[0025] Obtain the first decomposition feature of the behavioral data of each first account group in the account group within a preset period;

[0026] Obtain the second decomposition feature of the behavioral data in the second account group within a preset period;

[0027] Determine the maximum difference between the first decomposition feature and the second decomposition feature in each of the account groups, and select the account group with the smallest maximum difference among the multiple account groups as the target account group.

[0028] According to a second aspect of the present disclosure, an apparatus for determining account grouping is provided, comprising:

[0029] The extraction module is used to randomly extract a preset number of buckets from the bucket set to obtain candidate account groups. Each bucket includes multiple accounts. The bucket set is obtained by bucketing the accounts based on their identification information.

[0030] The first acquisition module is used to acquire the first behavior data of the accounts in the candidate account group and the second behavior data of the accounts in the remaining buckets, wherein the remaining buckets are the buckets in the bucket set other than the candidate account group.

[0031] The first determining module is used to determine the decomposition features of the first behavioral data and the second behavioral data within a preset time period, respectively.

[0032] The second determining module is used to determine a plurality of first account groups from a plurality of candidate account groups based on the decomposition features.

[0033] In one possible implementation, the first acquisition module includes:

[0034] The first acquisition submodule is used to acquire the mean and variance of the behavioral data of accounts in the bucket set within a preset time period;

[0035] The processing submodule is used to standardize the behavioral data based on the mean and the variance to obtain standard behavioral data.

[0036] The second acquisition submodule is used to acquire the first behavior data of the accounts in the candidate account group and the second behavior data of the accounts in the remaining buckets from the standard behavior data.

[0037] In one possible implementation, the second determining module includes:

[0038] The third acquisition submodule is used to acquire the difference between the decomposition features of the first behavioral data within a preset time period and the decomposition features of the second behavioral data within a preset time period, and to determine the candidate account group with the smallest preset number of differences.

[0039] The fourth acquisition submodule is used to acquire the sliding average of the first row data of each candidate account group within a preset period, and to group the candidate account group with the smallest sliding average as a first account group.

[0040] An update submodule is used to remove the buckets of the account groups from the bucket set in order to update the bucket set;

[0041] The generation submodule is used to repeatedly execute the following based on the updated bucket set: obtaining the difference between the decomposition features of the first behavior data and the second behavior data within a preset time period, determining the candidate account group, obtaining the sliding average of the first behavior data of the candidate account within a preset period, and obtaining the next first account group, until the number of the obtained first account groups reaches a preset value.

[0042] In one possible implementation, the second determining module includes:

[0043] The first determining submodule is used to determine multiple initial first account groups from multiple candidate account groups based on the decomposition features;

[0044] The verification submodule is used to determine the significance of the difference between any two initial first account groups among the plurality of initial first account groups. If no two initial first account groups are significant, then the plurality of initial first account groups are taken as the first account group.

[0045] In one possible implementation, the device further includes:

[0046] The second acquisition module is used to acquire multiple account groups and to select a first account group from the multiple account groups as a second account group, wherein each account group includes the multiple first account groups;

[0047] The second acquisition module is used to acquire the first decomposition feature of the behavioral data of each first account group in the account group within a preset period;

[0048] The third acquisition module is used to acquire the second decomposition features of the behavioral data in the second account group within a preset period;

[0049] The third determining module is used to determine the maximum difference between the first decomposition feature and the second decomposition feature in each of the account groups, and to take the account group with the smallest maximum difference among the multiple account groups as the target account group.

[0050] According to a third aspect of the present disclosure, an electronic device is provided, comprising:

[0051] processor;

[0052] Memory used to store the processor's executable instructions;

[0053] The processor is configured to execute the instructions to implement the account grouping determination method according to any one of the embodiments of this disclosure.

[0054] According to a fourth aspect of the present disclosure, a computer-readable storage medium is provided that, when instructions in the computer-readable storage medium are executed by a processor of an electronic device, enables the electronic device to perform the account grouping determination method according to any one of the embodiments of the present disclosure.

[0055] According to a fifth aspect of the present disclosure, a computer program product is provided, comprising instructions that, when executed by a processor of an electronic device, enable the electronic device to perform an account grouping determination method according to any one of the embodiments of the present disclosure.

[0056] The technical solution provided by the embodiments of this disclosure brings at least the following beneficial effects: In the embodiments of this disclosure, the decomposition characteristics of the first behavioral data within a preset time period and the decomposition characteristics of the second behavioral data within a preset time period are used as filtering conditions, and the filtered account groups can have the technical effect of minimal difference. When these account groups are used for scheme comparison testing in the future, the differences between the account groups are minimal, which ensures the accuracy of the scheme comparison test.

[0057] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0058] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure, and are not intended to unduly limit this disclosure.

[0059] Figure 1 This is a flowchart of a method for determining account grouping in the prior art.

[0060] Figure 2 This is a flowchart illustrating a method for determining account groups according to an exemplary embodiment.

[0061] Figure 3 This is a flowchart illustrating a method for determining account groups according to an exemplary embodiment.

[0062] Figure 4 This is a schematic block diagram illustrating an account grouping determination device according to an exemplary embodiment.

[0063] Figure 5 This is a schematic block diagram of an electronic device according to an exemplary embodiment.

[0064] Figure 6 This is a schematic block diagram of a computer program product according to an exemplary embodiment. Detailed Implementation

[0065] To enable those skilled in the art to better understand the technical solutions of this disclosure, the technical solutions in the embodiments of this disclosure will be clearly and completely described below with reference to the accompanying drawings.

[0066] It should be noted that the terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this disclosure are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this disclosure described herein can be implemented in orders other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this disclosure. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this disclosure as detailed in the appended claims.

[0067] It should also be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for display, data used for analysis, etc.) involved in this disclosure are all information and data authorized by the user or fully authorized by all parties.

[0068] Figure 2 This is a flowchart illustrating a method for determining account groups according to an exemplary embodiment. (Reference) Figure 2 As shown, the method is applied to a terminal or server and includes the following steps:

[0069] Step S201: Randomly select a preset number of sub-buckets from the bucket set to obtain candidate account groups. Each sub-bucket includes multiple accounts. The bucket set is obtained by sub-bucketing accounts based on their identification information.

[0070] In this embodiment of the disclosure, the bucket set includes multiple buckets obtained after bucketing all candidate users. Random, non-replacement sampling is performed on the buckets in the bucket set to obtain candidate account groups. Each candidate account group includes a preset number of buckets, and each bucket includes multiple accounts. In one example, before randomly selecting a preset number of buckets from the bucket set to obtain candidate groups, the method further includes: bucketing the candidate accounts based on the account's identification information to obtain the bucket set. In this embodiment of the disclosure, a hash function can be used to bucket the candidate accounts, for example, dividing the candidate accounts into 1000 buckets. Each bucket number can be represented as Hash(World_name + Uid) % 1000, where Uid represents the account's identification information, and World_name represents a specific application platform.

[0071] Step S203: Obtain the first line data of the accounts in the candidate account group and the second line data of the accounts in the remaining buckets, wherein the remaining buckets are the buckets in the bucket set other than the candidate account group.

[0072] In this embodiment of the disclosure, the account behavior data may include user operation behavior data, such as click behavior, download behavior, favorite behavior, forwarding behavior, etc. Behavior data from all buckets in the bucket set can be collected, and from the behavior data, first behavior data for candidate account groups and second behavior data for the remaining buckets can be selected.

[0073] Step S205: Determine the decomposition features of the first behavioral data and the second behavioral data within a preset time period.

[0074] In this embodiment of the disclosure, the first behavioral data and the second behavioral data may include various types of behavioral data, such as likes, purchases, and reviews. In one example, the behavioral data (including the first and second behavioral data) within a preset time period are decomposed to obtain corresponding decomposition features. The decomposition method may include STL decomposition, X11 or X12 decomposition, etc. The decomposition features may include mean features, seasonal features, periodic features, and volatility features, etc. In one example, the mean of the first behavioral data within the preset time period can be obtained as follows: within the preset time period, such as 30 days, 60 days, 90 days, etc., the first behavioral data for each day is accumulated, and the ratio of the accumulated sum to the number of days is used as the mean feature of the first behavioral data within the preset time period. Similarly, the mean feature of the second behavioral data within the preset time period can also be determined by the ratio of the accumulated sum of the behavioral data within the preset time period to the preset time period. In another example, the mean features of the first and second behavioral data within the preset time period can be obtained by accumulating and averaging the behavioral data of the same type, respectively, according to the type of behavioral data.

[0075] Step S207: Determine multiple first account groups from multiple candidate account groups based on the decomposition features.

[0076] In one example of this disclosure, the account group with the smallest difference between the decomposition features of the first row of data and the decomposition features of the second row of data can be selected from multiple candidate account groups. The buckets already identified as account groups in the bucket set are removed to update the buckets in the bucket set. Steps 301, 303, and 305 are repeated from the updated set to obtain the next account group. This process continues until the number of account groups reaches a preset requirement, at which point the resulting multiple account groups are used as experimental account groups. In another example, multiple account groups with the smallest difference between the decomposition features of the first row of data and the decomposition features of the second row of data can be selected from multiple candidate account groups. Then, the time series stability of the multiple account groups is evaluated to select account groups with more stable time series. The buckets already identified as account groups in the bucket set are removed to update the buckets in the bucket set. Steps 201, 203, and 205 are repeated from the updated set to obtain the next account group. This process continues until the number of account groups reaches a preset requirement, at which point the resulting multiple account groups are used as experimental account groups.

[0077] In this embodiment of the disclosure, considering that the change in bucket behavior data over time is a non-stationary process, it is represented as dX. i,t =μ t dt+σ t dW t i is the bin number, t is the time variable, and X is the time variable. i,t μ represents the behavioral data of bucket i on day t. t σ represents the mean of the data for this action. t W represents the standard deviation of the behavioral data. t Representing Brownian motion. The mean characteristic of the bucketed behavioral data can be represented as: Where N represents the number of buckets, X i,t This represents the behavioral data of bucket i on day t. The mean feature... obey Where, μ t This represents the mean of the behavioral data. Therefore, the difference between the bucketed behavioral data and the mean is approximately a stationary process, expressed as... The expected value of the difference described in this process is ε iThe difference between the behavioral data of each bucket and the mean is a fixed effect. Since the stochastic process of the difference between the bucket and the mean is stationary, buckets that have historically exceeded the mean can be combined with buckets that have fallen below the mean to create account groups. This minimizes the total difference between the account groups and the mean, thereby achieving the goal of minimizing the difference between each account group.

[0078] In this embodiment of the present disclosure, the decomposition characteristics of the first and second behavioral data within a preset time period are used as filtering conditions to ensure that the filtered account groups have minimal differences. In subsequent scheme comparison tests using these account groups, the differences between the account groups are minimal, thus ensuring the accuracy of the scheme comparison tests.

[0079] In one possible implementation, obtaining the first behavioral data of accounts in the candidate account group and the second behavioral data of accounts in the remaining buckets includes:

[0080] Obtain the mean and variance of the behavioral data of accounts in the bucket set over a preset time period;

[0081] The original behavioral data is standardized based on the mean and variance to obtain standardized behavioral data.

[0082] Obtain the first behavioral data of the accounts in the candidate account group and the second behavioral data of the accounts in the remaining buckets from the standard behavioral data.

[0083] In this embodiment of the disclosure, the mean μ of the behavioral data of accounts in the bucket set over a preset time period is... m It can be represented as x n,t This represents the raw behavioral data of the nth bucket at time t. The variance σ of the behavioral data of the accounts in the bucket set over a preset time period is also represented. m It can be represented as μ m and σ m In the expression, N represents the number of buckets in the bucket set, T represents the preset time period, and m represents the type of data in the first row. Let the standard behavior data of the nth bucket at time t be represented. Then, the first behavioral data of the candidate account group and the second behavioral data of the accounts in the remaining buckets are obtained from the standard behavioral data.

[0084] This embodiment utilizes the mean and variance of the original behavioral data within a preset time period to standardize the original behavioral data, thereby obtaining dimensionless behavioral data. This eliminates the data incompatibility problem caused by different units of different types of behavioral parameters, and enables the processing of various types of behavioral data.

[0085] In one possible implementation, determining multiple first account groups from multiple candidate account groups based on the decomposition features includes:

[0086] Obtain the difference between the decomposition features of the first behavioral data within a preset time period and the decomposition features of the second behavioral data within a preset time period, and determine a preset number of candidate account groups with the smallest difference.

[0087] In this embodiment, the methods for obtaining the decomposition features of the first behavioral data within a preset time period and the decomposition features of the second behavioral data within a preset time period are the same as in the above embodiments, and will not be repeated here. From a randomly selected group of candidate accounts, a preset number of candidate account groups with the smallest differences are determined. For example, from 40,000 candidate account groups, the 10 candidate account groups with the smallest differences are determined.

[0088] Obtain the sliding average of the first row data of each candidate account group within a preset period, and select the candidate account group with the smallest sliding average as the first account group in the experimental account group.

[0089] In this embodiment, the moving average is a tool for analyzing time series data, used to describe the stationarity of data over a certain period. A moving average model can be used to obtain the moving average of each behavioral data point within a preset period. For example, in this embodiment, the mean of the first behavioral data point within a preset time period can be expressed as: Among the sub-buckets in candidate group B, (x n,t ) m Let represent the user behavior data of the m-th type of user behavior in the n-th bucket at time t, and size(B) represent the number of buckets in the candidate group. The mean of the second behavior data over the preset time period can be expressed as: Among them, (x n,t ) m Let N represent the user behavior data for the m-th type of user behavior in the n-th bucket at time t, where N represents the number of buckets in the bucket set. The difference between the two means can be expressed as (diff) B,t ) m ,but The difference is input into a moving average model, and a preset period is set, for example, 7 days, to obtain the moving average of the candidate groups. The one with the smallest moving average is selected as one of the first accounts in the group.

[0090] Remove the buckets for the account group from the bucket set to update the bucket set.

[0091] Based on the updated bucket set, the next account group is obtained until the number of account groups obtained reaches a preset value. The multiple account groups obtained are then used as experimental account groups.

[0092] In this embodiment of the disclosure, the buckets of the account groups are removed from the bucket set and the bucket set is updated. Based on the updated bucket set, the process of obtaining the difference between the decomposition features of the first behavioral data and the second behavioral data within a preset time period, determining the candidate account group, obtaining the sliding average of the first behavioral data of the candidate account within a preset period, and obtaining the next first account group is repeated until the number of first account groups reaches a preset value.

[0093] In this embodiment of the disclosure, by using the mean of the first behavioral data within a preset time period and the decomposition characteristics of the second behavioral data within a preset time period as screening conditions, a preset number of candidate account groups with the smallest difference between the two decomposition characteristics are determined. At this point, the differences between the candidate account groups are already very small. Furthermore, by comparing the moving average of each candidate account group within a preset period, which can reflect the stability of the behavioral data within the preset period, the candidate account group with the smallest moving average is selected as the first account group, that is, the one with the most stable sequence is selected as the first account group.

[0094] The embodiments disclosed herein can obtain account groups with less difference and higher temporal stability.

[0095] In one possible implementation, determining multiple first account groups from multiple candidate account groups based on the decomposition features includes:

[0096] Multiple initial first account groups are determined from multiple candidate account groups based on the decomposition features;

[0097] Determine the significance of the difference between any two initial first account groups among the plurality of initial first account groups. If no two initial first account groups are significant, then the plurality of initial first account groups are taken as the first account group.

[0098] In this embodiment of the disclosure, multiple account groups can be determined from multiple candidate account groups according to any of the methods described in the above embodiments, based on the decomposition features. A specific method for determining the significance of the difference between any two initial first account groups among the multiple account groups can utilize a significance test. First, it is assumed that the data distribution of the initial first account groups is not different. If the probability (P-value) of drawing a sample from the initial first account group is less than the significance level α, then a low-probability event has occurred, and the null hypothesis is rejected, meaning there is a distributional difference among the multiple initial first account groups. If the probability of the sample occurring is greater than the significance level, then the null hypothesis is accepted, and it is considered that there is no distributional difference among the multiple account groups. Here, the significance level α refers to events with a probability less than 0.05 being considered low-probability events. Of course, it can be set to 0.1 or 0.001 depending on the actual situation. Statistically, this probability is called the significance level α. For example, sample accounts are drawn from multiple account groups, and the differences in sample account behavior data from different account groups are compared. If the probability of no difference is less than the significance level α, it means that the original hypothesis that there is no difference between account groups is not true. Conversely, if the probability of a difference is greater than the significance level α, it means that the original hypothesis that there is no difference between account groups is true, that is, no two initial first account groups are significant.

[0099] In this embodiment of the disclosure, by performing saliency verification on multiple account groups, it can be ensured that the obtained account groups are not saliency to each other, thereby ensuring the accuracy of subsequent scheme comparison tests.

[0100] In one possible implementation, the method for determining the account groups further includes:

[0101] Obtain multiple account groups, and select one first account group from the multiple account groups as a second account group, wherein each account group includes the multiple first account groups;

[0102] Obtain the first decomposition feature of the behavioral data of each first account group in the account group within a preset period;

[0103] Obtain the second decomposition feature of the behavioral data in the account control group within a preset period;

[0104] Determine the maximum difference between the first decomposition feature and the second decomposition feature in each of the experimental account groups, and select the account group with the smallest maximum difference among the multiple account groups as the target account group.

[0105] In this embodiment of the disclosure, multiple account groups are obtained according to any of the methods in the above embodiments, such as account group A, account group B, and account group C. Account group A includes: first account group (1), first account group (2), and first account group (3); account group B includes: first account group (4), first account group (5), and first account group (6); and account group C includes: first account group (7), first account group (8), and first account group (9). In one example, any one of the first account group (1) to the first account group (9) can be selected as the second account group, for example, the first account group (9) can be selected as the second account group. The first mean of the behavioral data of each first account group in each account group is obtained within a preset period; the second mean of the behavioral data in the second account group (such as the first account group (9)) is obtained within a preset period. The maximum difference between the first and second means in each experimental account group is determined. Taking account group A as an example, the difference between account group (1) and the account control group is, for example, 0.2; the difference between account group (2) and the account control group is, for example, 0.3; and the difference between account group (3) and the account control group is, for example, 0.1. Therefore, the maximum difference between the first and second means in account group A is 0.3. Similarly, the maximum difference between the first and second means in account group B is 0.25, and the maximum difference between the first and second means in account group C is 0.13. The account group with the smallest maximum difference is selected. In this embodiment, the maximum difference of account group C (0.13) is the smallest among the three account groups, and this account group is selected as the target account group.

[0106] In this embodiment of the disclosure, by repeating the method for determining account groups in the above embodiments, multiple sets of account groups are obtained. The mean difference between the first account group and the second account group in the multiple sets of account groups is compared. First, the maximum difference between each first account group and the second account group in each account group is determined. Then, the account group with the smallest maximum difference is selected from the multiple account groups. This ensures that the difference between each first account group in the selected account group and the account control group is small. The final account group is selected in sequence, and the difference between each first account group in the account group is the smallest.

[0107] Figure 3 This is a flowchart illustrating a method for determining account groups according to an exemplary embodiment. (Reference) Figure 3 As shown, the method for determining account groups in this embodiment of the disclosure is as follows:

[0108] Obtaining available bucket metrics data corresponds to the steps in the above embodiment: randomly selecting a preset number of buckets from the bucket set to obtain candidate account groups, each bucket including multiple accounts, the bucket set being obtained by bucketing accounts based on their identification information; obtaining the first behavioral data of accounts in the candidate account groups and the second behavioral data of accounts in the remaining buckets, the remaining buckets being the buckets in the bucket set excluding the candidate account groups. Available buckets are those accounts in the bucket set not yet identified as account groups. Specifically, each time an account group is determined, the buckets of that account group are removed from the bucket set to update the bucket set. Metric data may include user account behavioral data, such as likes, shares, and reads.

[0109] Figure 3 In this process, the indicator data is standardized, corresponding to the step of obtaining the mean and variance of the behavioral data of the accounts in the bucket set within a preset time period in the above embodiment; based on the mean and the variance, the behavioral data is standardized to obtain standard behavioral data. Figure 2 The group information includes group traffic and group type. Group traffic refers to the number of website visits, reflected in the number of accounts accessing the website. For example, if the required group traffic is 10% of the total website visits, and there are 1000 buckets in the bucket set, then the group size is 1000 × 10% (100) buckets, meaning the size of the first account group is 100 buckets. The number of the first account group and the number of buckets in the account group are preset according to the requirements of the scheme comparison experiment.

[0110] Figure 3 In this process, minimizing the bucket set loss function and iteratively selecting groups corresponds to the steps in the above embodiment: obtaining the difference between the decomposition features of the first behavioral data and the second behavioral data within a preset time period, determining a preset number of candidate account groups with the smallest difference; obtaining the moving average of the first behavioral data of each candidate account group within a preset period, and taking the candidate account group with the smallest moving average as a first account group; removing the buckets of the account groups from the bucket set to update the bucket set; based on the updated bucket set, obtaining the difference between the decomposition features of the first behavioral data and the second behavioral data within a preset time period, determining candidate account groups, and obtaining the moving average of the first behavioral data of the candidate accounts within a preset period to obtain the next first account group, until the number of obtained first account groups reaches a preset value, and taking the multiple obtained first account groups as an account group.

[0111] Figure 3In this process, the first group is selected as the baseline group, and the significance of other groups is determined. This corresponds to the step in the above embodiment where multiple initial first account groups are determined from multiple candidate account groups based on the decomposition. The significance of the difference between any two initial first account groups is determined. If no two initial first account groups are significant, then the multiple initial first account groups are used as account groups. If there is a significant difference between two initial first account groups, the process of obtaining account groups fails and needs to be re-screened.

[0112] Table 1

[0113]

[0114] In this embodiment, 1200 simulations are performed using a 30-day in-sample training set for different traffic volumes (0.5%, 1%, 2%, 3%, 5%, 10%, 20%, 25%). Each optimization involves two experimental groups with the same traffic volume (e.g., for 5% traffic, two 5% experimental groups are set up for optimization each time). The mean of the false positive rate for the main station, main station duration, express version DAU, express version duration, and the group-level false positive rate is calculated for each optimized experimental group over a 7-day interval (0, 7, 14, 30, 60 days after optimization). The difference between this value and the false positive rate under the random traffic splitting strategy is used to measure the effectiveness of the traffic splitting strategy.

[0115] The overall results are shown in Table 1. The false positive rate obtained through offline triage is significantly lower than that of the current random triage method. This demonstrates that the triage strategy can maintain the difference between experimental groups less than that under random triage within 60 days out of the sample. If the experimental length is within 60 days, the effect of offline triage in reducing the false positive rate is approximately 4.7%.

[0116] For the effects of different traffic volumes, see Table 2. The offline traffic splitting strategy ensures that the false positive rate is significantly lower than that of the random traffic splitting method within 60 days, and does not exceed 5% across all traffic volumes. The smaller the traffic volume, the greater the effect of offline traffic splitting. For 0.5% of the traffic, offline traffic splitting reduces the false positive rate by at least 8%.

[0117] Table 2

[0118]

[0119] It should be understood that, although Figure 2 and Figure 3 The steps in the flowchart are shown sequentially as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order in which these steps are executed, and they can be performed in other orders. Figure 2 and Figure 3 At least some of the steps in the process may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but may be executed at different times. The execution order of these steps or stages is not necessarily sequential, but may be executed in turn or alternately with other steps or at least some of the steps or stages in other steps.

[0120] It is understood that the same / similar parts between the various embodiments of the methods described above in this specification can be referred to each other. Each embodiment focuses on the differences from other embodiments, and relevant parts can be referred to the description of other method embodiments.

[0121] Figure 4 This is a schematic block diagram illustrating an account grouping determination device according to an exemplary embodiment. (Refer to...) Figure 4 The device includes:

[0122] Extraction module 401 is used to randomly extract a preset number of sub-buckets from the bucket set to obtain candidate account groups. Each sub-bucket includes multiple accounts. The bucket set is obtained by sub-bucketing accounts based on their identification information.

[0123] The first acquisition module 402 is used to acquire the first behavior data of the accounts in the candidate account group and the second behavior data of the accounts in the remaining buckets, wherein the remaining buckets are the buckets in the bucket set other than the candidate account group.

[0124] The first determining module 403 is used to determine the decomposition features of the first behavioral data and the second behavioral data within a preset time period;

[0125] The second determining module 404 is used to determine a plurality of first account groups from a plurality of candidate account groups based on the decomposition features.

[0126] In one possible implementation, the first acquisition module includes:

[0127] The first acquisition submodule is used to acquire the mean and variance of the behavioral data of accounts in the bucket set within a preset time period;

[0128] The processing submodule is used to standardize the behavioral data based on the mean and the variance to obtain standard behavioral data.

[0129] The second acquisition submodule is used to acquire the first behavior data of the accounts in the candidate account group and the second behavior data of the accounts in the remaining buckets from the standard behavior data.

[0130] In one possible implementation, the second determining module includes:

[0131] The third acquisition submodule is used to acquire the difference between the decomposition features of the first behavioral data within a preset time period and the decomposition features of the second behavioral data within a preset time period, and to determine the candidate account group with the smallest preset number of differences.

[0132] The fourth acquisition submodule is used to acquire the sliding average of the first row data of each candidate account group within a preset period, and to group the candidate account group with the smallest sliding average as a first account group.

[0133] An update submodule is used to remove the buckets of the account groups from the bucket set in order to update the bucket set;

[0134] The generation submodule is used to repeatedly execute the following based on the updated bucket set: obtaining the difference between the decomposition features of the first behavior data and the second behavior data within a preset time period, determining the candidate account group, obtaining the sliding average of the first behavior data of the candidate account within a preset period, and obtaining the next first account group, until the number of the obtained first account groups reaches a preset value.

[0135] In one possible implementation, the second determining module includes:

[0136] The first determining submodule is used to determine multiple initial first account groups from multiple candidate account groups based on the decomposition features;

[0137] The verification submodule is used to determine the significance of the difference between any two initial first account groups among the plurality of initial first account groups. If no two initial first account groups are significant, then the plurality of initial first account groups are taken as the first account group.

[0138] In one possible implementation, the device further includes:

[0139] The second acquisition module is used to acquire multiple account groups and to select a first account group from the multiple account groups as a second account group, wherein each account group includes the multiple first account groups;

[0140] The second acquisition module is used to acquire the first decomposition feature of the behavioral data of each first account group in the account group within a preset period;

[0141] The third acquisition module is used to acquire the second decomposition features of the behavioral data in the second account group within a preset period;

[0142] The third determining module is used to determine the maximum difference between the first decomposition feature and the second decomposition feature in each of the account groups, and to take the account group with the smallest maximum difference among the multiple account groups as the target account group.

[0143] Regarding the apparatus in the above embodiments, the specific manner in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated upon here.

[0144] Figure 5 This is a block diagram illustrating an electronic device 500 according to an exemplary embodiment. For example, the electronic device 500 may be a mobile phone, computer, digital broadcasting terminal, messaging device, game console, tablet device, medical device, fitness equipment, personal digital assistant, etc.

[0145] Reference Figure 5 The electronic device 500 may include one or more of the following components: processing component 502, memory 504, power supply component 506, multimedia component 508, audio component 510, input / output (I / O) interface 512, sensor component 514, and communication component 516.

[0146] Processing component 502 typically controls the overall operation of electronic device 500, such as operations associated with display, telephone calls, data communication, camera operation, and recording operations. Processing component 502 may include one or more processors 520 to execute instructions to complete all or part of the steps of the methods described above. Furthermore, processing component 502 may include one or more modules to facilitate interaction between processing component 502 and other components. For example, processing component 502 may include a multimedia module to facilitate interaction between multimedia component 508 and processing component 502.

[0147] Memory 504 is configured to store various types of data to support the operation of electronic device 500. Examples of such data include instructions for any application or method operating on electronic device 500, contact data, phonebook data, messages, pictures, videos, etc. Memory 504 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, optical disk, or graphene storage.

[0148] Power supply component 506 provides power to various components of electronic device 500. Power supply component 506 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power to electronic device 500.

[0149] Multimedia component 508 includes a screen that provides an output interface between the electronic device 500 and the user. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen may be implemented as a touchscreen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touches, swipes, and gestures on the touch panel. The touch sensors may sense not only the boundaries of the touch or swipe action but also the duration and pressure associated with the touch or swipe operation. In some embodiments, multimedia component 508 includes a front-facing camera and / or a rear-facing camera. When the electronic device 500 is in an operating mode, such as a shooting mode or a video mode, the front-facing camera and / or the rear-facing camera may receive external multimedia data. Each front-facing camera and rear-facing camera may be a fixed optical lens system or have focal length and optical zoom capabilities.

[0150] Audio component 510 is configured to output and / or input audio signals. For example, audio component 510 includes a microphone (MIC) configured to receive external audio signals when electronic device 500 is in an operating mode, such as call mode, recording mode, and voice recognition mode. The received audio signals may be further stored in memory 504 or transmitted via communication component 516. In some embodiments, audio component 510 also includes a speaker for outputting audio signals.

[0151] I / O interface 512 provides an interface between processing component 502 and peripheral interface modules, such as keyboards, click wheels, buttons, etc. These buttons may include, but are not limited to, home buttons, volume buttons, power buttons, and lock buttons.

[0152] Sensor assembly 514 includes one or more sensors for providing state assessments of various aspects of electronic device 500. For example, sensor assembly 514 can detect the on / off state of electronic device 500, the relative positioning of components such as the display and keypad of electronic device 500, changes in position of electronic device 500 or its components, the presence or absence of user contact with electronic device 500, orientation or acceleration / deceleration of device 500, and temperature changes of electronic device 500. Sensor assembly 514 may include a proximity sensor configured to detect the presence of nearby objects without any physical contact. Sensor assembly 514 may also include a light sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, sensor assembly 514 may also include an accelerometer, gyroscope, magnetometer, pressure sensor, or temperature sensor.

[0153] Communication component 516 is configured to facilitate wired or wireless communication between electronic device 500 and other devices. Electronic device 500 can access wireless networks based on communication standards, such as WiFi, carrier networks (such as 2G, 3G, 4G, or 5G), or combinations thereof. In one exemplary embodiment, communication component 516 receives broadcast signals or broadcast-related information from an external broadcast management system via a broadcast channel. In one exemplary embodiment, communication component 516 also includes a near-field communication (NFC) module to facilitate short-range communication. For example, the NFC module may be implemented based on radio frequency identification (RFID) technology, Infrared Data Association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.

[0154] In an exemplary embodiment, the electronic device 500 may be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the methods described above.

[0155] In an exemplary embodiment, a computer-readable storage medium including instructions is also provided, such as a memory 504 including instructions, which can be executed by a processor 520 of an electronic device 500 to perform the above-described method. For example, the computer-readable storage medium may be a ROM, random access memory (RAM), CD-ROM, magnetic tape, floppy disk, and optical data storage device, etc.

[0156] In an exemplary embodiment, a computer program product is also provided, which includes instructions that can be executed by a processor 520 of an electronic device 500 to perform the above-described method.

[0157] Figure 6 This is a block diagram illustrating an electronic device 600 according to an exemplary embodiment. For example, the electronic device 600 may be a server. (Refer to...) Figure 6 The electronic device 600 includes a processing component 620, which further includes one or more processors, and memory resources represented by memory 622 for storing instructions, such as application programs, that can be executed by the processing component 620. The application programs stored in memory 622 may include one or more modules, each corresponding to a set of instructions. Furthermore, the processing component 620 is configured to execute instructions to perform the methods described above.

[0158] Electronic device 600 may further include: a power supply component 624 configured to perform power management of electronic device 600, a wired or wireless network interface 626 configured to connect electronic device 600 to a network, and an input / output (I / O) interface 628. Electronic device 600 may operate on an operating system stored in memory 622, such as Windows Server, Mac OS X, Unix, Linux, FreeBSD, or similar.

[0159] In an exemplary embodiment, a computer-readable storage medium including instructions is also provided, such as a memory 622 including instructions, which can be executed by a processor of an electronic device 600 to perform the above-described method. The storage medium may be a computer-readable storage medium, such as a ROM, random access memory (RAM), CD-ROM, magnetic tape, floppy disk, and optical data storage device.

[0160] In an exemplary embodiment, a computer program product is also provided, the computer program product including instructions that can be executed by a processor of an electronic device 600 to perform the above-described method.

[0161] It should be noted that the above-mentioned apparatus, electronic equipment, computer-readable storage medium, computer program product, etc., may also include other implementation methods according to the description of the method embodiments. For specific implementation methods, please refer to the description of the relevant method embodiments, which will not be elaborated here.

[0162] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This disclosure is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this disclosure are indicated by the claims.

[0163] It should be understood that this disclosure is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this disclosure is limited only by the appended claims.

Claims

1. A method for determining account grouping, characterized in that, include: A preset number of buckets are randomly selected from the bucket set to obtain candidate account groups. Each bucket includes multiple accounts. The bucket set is obtained by bucketing accounts based on their identification information. Obtain the first row data of the accounts in the candidate account group and the second row data of the accounts in the remaining buckets, wherein the remaining buckets are the buckets in the bucket set other than the candidate account group; Determine the decomposition features of the first behavioral data and the second behavioral data within a preset time period, respectively; Multiple first account groups are determined from multiple candidate account groups based on the decomposition features; The step of determining multiple first account groups from multiple candidate account groups based on the decomposition features includes: obtaining the difference between the decomposition features of the first behavioral data within a preset time period and the decomposition features of the second behavioral data within a preset time period, and determining a preset number of candidate account groups with the smallest difference. Obtain the moving average of the first behavior data of each candidate account group within a preset period, and select the candidate account group with the smallest moving average as a first account group; remove the buckets of the account group from the bucket set to update the bucket set; based on the updated bucket set, repeat the following steps: obtain the difference between the decomposition features of the first behavior data and the second behavior data within a preset time period, determine the candidate account group, obtain the moving average of the first behavior data of the candidate account within a preset period, and obtain the next first account group, until the number of first account groups obtained reaches a preset value.

2. The method according to claim 1, characterized in that, The step of obtaining the first behavior data of accounts in the candidate account group and the second behavior data of accounts in the remaining buckets includes: Obtain the mean and variance of the behavioral data of accounts in the bucket set over a preset time period; The behavioral data is standardized based on the mean and variance to obtain standard behavioral data. Obtain the first behavioral data of the accounts in the candidate account group and the second behavioral data of the accounts in the remaining buckets from the standard behavioral data.

3. The method according to claim 1, characterized in that, The step of determining multiple first account groups from multiple candidate account groups based on the decomposition features includes: Multiple initial first account groups are determined from multiple candidate account groups based on the decomposition features; Determine the significance of the difference between any two initial first account groups among the plurality of initial first account groups. If no two initial first account groups are significant, then the plurality of initial first account groups are taken as the first account group.

4. The method according to any one of claims 1 to 3, characterized in that, The method further includes: Obtain multiple account groups, and select one first account group from the multiple account groups as a second account group, wherein each account group includes the multiple first account groups; Obtain the first decomposition feature of the behavioral data of each first account group in the account group within a preset period; Obtain the second decomposition feature of the behavioral data in the second account group within a preset period; Determine the maximum difference between the first decomposition feature and the second decomposition feature in each of the account groups, and select the account group with the smallest maximum difference among the multiple account groups as the target account group.

5. An account grouping determination device, characterized in that, include: The extraction module is used to randomly extract a preset number of buckets from the bucket set to obtain candidate account groups. Each bucket includes multiple accounts. The bucket set is obtained by bucketing the accounts based on their identification information. The first acquisition module is used to acquire the first behavior data of the accounts in the candidate account group and the second behavior data of the accounts in the remaining buckets, wherein the remaining buckets are the buckets in the bucket set other than the candidate account group. The first determining module is used to determine the decomposition features of the first behavioral data and the second behavioral data within a preset time period, respectively. The second determining module is used to determine a plurality of first account groups from a plurality of candidate account groups based on the decomposition features; The second determining module includes: a third obtaining submodule, used to obtain the difference between the decomposition features of the first behavioral data and the decomposition features of the second behavioral data within a preset time period, and determine a preset number of candidate account groups with the smallest difference; a fourth obtaining submodule, used to obtain the moving average of the first behavioral data of each candidate account group within a preset period, and take the candidate account group with the smallest moving average as a first account group; an updating submodule, used to remove the buckets of the account groups from the bucket set to update the bucket set; and a generating submodule, used to repeatedly execute, based on the updated bucket set: obtaining the difference between the decomposition features of the first behavioral data and the second behavioral data within a preset time period, determining candidate account groups, and obtaining the moving average of the first behavioral data of the candidate accounts within a preset period to obtain the next first account group, until the number of obtained first account groups reaches a preset value.

6. The apparatus according to claim 5, characterized in that, The first acquisition module includes: The first acquisition submodule is used to acquire the mean and variance of the behavioral data of accounts in the bucket set within a preset time period; The processing submodule is used to standardize the behavioral data based on the mean and the variance to obtain standard behavioral data. The second acquisition submodule is used to acquire the first behavior data of the accounts in the candidate account group and the second behavior data of the accounts in the remaining buckets from the standard behavior data.

7. The apparatus according to claim 5, characterized in that, The second determining module includes: The first determining submodule is used to determine multiple initial first account groups from multiple candidate account groups based on the decomposition features; The verification submodule is used to determine the significance of the difference between any two initial first account groups among the plurality of initial first account groups. If no two initial first account groups are significant, then the plurality of initial first account groups are taken as the first account group.

8. The apparatus according to any one of claims 5 to 7, characterized in that, The device further includes: The second acquisition module is used to acquire multiple account groups and to select a first account group from the multiple account groups as a second account group, wherein each account group includes the multiple first account groups; The second acquisition module is used to acquire the first decomposition feature of the behavioral data of each first account group in the account group within a preset period; The third acquisition module is used to acquire the second decomposition features of the behavioral data in the second account group within a preset period; The third determining module is used to determine the maximum difference between the first decomposition feature and the second decomposition feature in each of the account groups, and to take the account group with the smallest maximum difference among the multiple account groups as the target account group.

9. An electronic device, characterized in that, include: processor; Memory used to store the processor's executable instructions; The processor is configured to execute the instructions to implement the method for determining account groups as described in any one of claims 1 to 4.

10. A computer-readable storage medium, characterized in that, When the instructions in the computer-readable storage medium are executed by the processor of the electronic device, the electronic device is able to perform the method for determining account groups as described in any one of claims 1 to 4.

Images